Integrations and credentials

Integrations are catalog entries per workspace. Most connections store a workspace-scoped credential (encrypted) and mark the integration connected. That status records a saved account; it does not certify authentication or every permission. Segment and manually saved Google Tasks, CircleCI, Brandfetch, Action Network, Affinity, Adalo, APITemplate.io, Airtop and Autopilot Journeys accounts explicitly show saved · unverified. Secrets you type in Settings → Secrets are the same credential records.

Personal browser connections keep a separate browser profile for each account and website. Use Manage my browser connections to enter the site's login page, account page and identifying account text, then sign in and save. The Web browser blocks verify access, report connection status or read a page using that session. To let your agent browse, enable Allow my agent to read on your connection. For interactive browsing in Codex CLI or Claude Code, attach Playwright MCP · saved browser on the node and authorize its separate control grant. Existing MCP tools and skills remain attached. Shared workspace credentials do not grant access to someone else's browser. See workspace settings.

A personal browser connection is a session you signed in to yourself, not automatic sign-in: when the website expires it, sign in again. It serves your own agent and workflows only. Customers of a product you build use per-customer website logins instead (End users and memory). Which of these you can use depends on your deployment and your role.

After returning from OAuth in a workflow, open Editor → Resources, select the connected account for the pending node and choose Use this account. You can also use the warning's Open node button to select it in block configuration. A connected workspace account and a selected node account are separate states. For Production or a named preview, save the account under Connections for this destination as well; publication notices link to those choices.

Settings surfaces

The catalog search sits directly above Available integrations, below the CLI subscription connections. GitHub workflow review setup has its own Settings → Workflow reviews section. A GitHub account connection in Integrations still authorizes GitHub workflow blocks.

The app uses tRPC settings.listIntegrations / updateIntegration and settings.listCredentials / createCredential (also session REST /api/integrations and /api/credentials).

Webhook endpoints

In Settings → Integrations, the Webhook endpoints panel lists incoming event addresses. Search by workflow, trigger, method or URL, and filter All, Enabled or Disabled. Each card shows its workflow, trigger when available, creation date, HTTP method and status. Select the workflow name to open its canvas. Copy URL copies the complete address; you can also select the address manually.

Authorized members can Enable or Disable an endpoint. Test sends a sample event for the workflow; it is unavailable while the endpoint is disabled. The card shows test feedback and keeps the saved status if an update fails. Search and filters change only the displayed list. Members whose access does not include invocation URLs cannot copy or test those addresses.

A successful webhook test may return an accepted event before its queued run starts. Open Executions to follow that run. Waiting for usage capacity is shown separately from a failed delivery. A workflow with a custom HTTP response can acknowledge delivery without returning a run object.

Meeting reminders workflow template

In Home → Workflow templates → Productivity, choose Multi-calendar meeting reminders and Use template. This creates an editable draft workflow and a linked setup walkthrough. It does not start sending reminders.

1. Connect each account in Settings → Integrations. Use Google Calendar for Google/Gmail calendars and Microsoft Outlook for Outlook calendars; Gmail mail access alone does not grant calendar access. Choose Add another and distinct connection names for additional accounts.
2. Open the workflow. In each Google Calendar meetings or Outlook meetings block, select its workspace credential, calendar ID (primary selects the default calendar), and reminderMinutes (10 by default). Duplicate a calendar block and connect it before Combine meeting reminders for each additional account or calendar. Remove unused source blocks and reconnect their neighbors.
3. In Reminder settings, set the display timeZone to an IANA timezone, such as America/Mexico_City. Keep the shared asOf expression for manual tests. Scheduled runs use their claimed input.scheduledFor timestamp, so queue delays do not move an existing tick into a different minute.
4. Send reminder · replace this channel is a Slack example. Set its account and channel, or replace it with a supported email, Telegram, Teams or other notification block. Connect the condition's true branch to it. The sample Slack block uses {{outputs.digest.slackText}} to escape mention syntax in meeting titles. For other plain-text destinations, use {{outputs.digest.text}} (or {{outputs.digest.meetings}} for your own formatting). Connect the send result to Reminder result. The false branch sends nothing.
5. Test with a dedicated meeting and destination, inspect Executions, then deploy and enable the workflow's schedule. Keep Check every minute at * * * * *. Normal schedules run while the browser is closed; disable the schedule to stop future reminders. Deployment review and destination authorization still apply.

The read blocks follow provider pagination and normalize timed occurrences to UTC. All-day, cancelled and declined meetings are excluded. The collector combines entries only when their reported invitation IDs and start times match within that run. Google and Outlook can report different IDs for copies of the same invitation, especially recurring meetings; those copies remain separate entries. This template does not normalize cross-provider invitation identity. Each minute queries a distinct one-minute start-time slot, offset by the chosen notice time. Slot alignment can select a meeting less than one minute before its exact notice time; queueing and delivery can add delay. With the default source error settings, a failed provider read fails the workflow before notification; the run exposes the error. If you choose continue-on-error on a source, the digest contains only successful sources. Intervals skipped between claimed schedule ticks are not backfilled. Queued runs keep their claimed time and can send late. Manually repeating a run can repeat its notification; this template does not provide an exactly-once ledger across separate runs. Provider changes take effect on the next read.

These are normal workspace credentials and workflows, not private personal-calendar connections. Authorized workspace members can access workflow configuration, meeting output and execution history according to their role. Use a separate workspace when calendars should not be shared. Never put OAuth tokens or bot tokens into the template or message fields. Existing operator live controls and provider consent requirements still apply.

Catalog and connection

In Settings → Integrations, search for a service and choose Connect account on its card. For another account of the same service, choose Add another. Enter a Connection name, then use the provider sign-in button or the credential fields shown there. Trello, Freshdesk, Zendesk, Supabase, Stripe and Mailchimp show named credential fields here and in the workflow node’s Add account dialog. Structured accounts also offer Advanced JSON. Completed credential fields reappear if you temporarily filter that service out of the catalog and show it again. Adding another account starts with empty secret fields. The catalog uses the same service icons as workflow blocks. Connection errors appear below the actions; dismiss an error with its close button or retry sign-in.

Segment uses Save account and does not automatically test its Tracking key. No harmless authenticated read is available for this key profile. Its account stays explicitly unverified; bulk connection tests skip it without counting a pass or a failure. Saving or selecting that account does not send a sample event.

GET /api/v1/integrations returns { integrations[] } for the API key's workspace: key, name, category, description, status (available or connected), optional credentialId / credentialName / credentialType, credentialConfigured, connectedCredentials[], timestamps.

PATCH with { key, connected } upserts the connection row. Unknown keys return 400 "Unknown integration.".

When connected is true and credentialValue is present, the server creates or updates a workspace credential (type derived from the integration) and returns { integration, credential } with maskedValue (never the plaintext). Multiple named credentials can exist for one key (for example two Gmail accounts).

Supported OAuth keys include github, google-workspace, google-drive, gmail, jira, microsoft-teams, and others in integrationOAuthProviders. Start returns { authorizationUrl, provider, credentialName, expiresAt } and sets an OAuth state cookie, or 200 with missingConnectorSetup / mode: "manual" when client env is incomplete. Callback writes the token into a credential and redirects to returnTo (default Settings → Integrations). returnTo must be a safe internal path.

If Sign in with Google or another OAuth sign-in requires verification, Confluye opens the Verify MFA dialog in Integrations and explains which account connection needs it. Complete authenticator verification or setup there; Confluye retries that same connection with the selected credential name and opens the provider for consent. Canceling leaves the account unconnected and does not retry. Rejected codes stay in the dialog for correction. If the connection is still refused after verification, its error appears beside the sign-in button. An OAuth ready label describes the connector configuration; session verification and provider consent can still be required.

Notion, Supabase and Pinecone workflow nodes

Connect an account in Integrations, use Test connection, then select that saved account in the workflow block. Tokens remain in Credentials; do not put them in block inputs.

Notion uses the 2022-06-28 API contract, including its database operations. Share each required page/database with the integration and grant its required read/write capabilities. Search and list operations return one page, including has_more and next_cursor; pass the cursor as Start cursor for another page. This does not include the newer data-source API.

Supabase defaults to the project's public REST schema and uses the selected key's database permissions and row policies. Set Schema for another schema already exposed in the project's API settings. Filters are a JSON object of PostgREST column predicates (simple column names), for example {"status":"eq.open"}. Update/delete require at least one column predicate; query options such as limit or an empty logical group do not satisfy this requirement. List rows defaults to 50 rows; use Offset for another page. Upsert rows can specify On conflict for a unique column. Calling a database function can change data and is treated as a write.

Pinecone operations manage indexes; vector insertion, retrieval and similarity search are not included. Configuration/deletion requests can be accepted before the change completes: check the index again before depending on its new state. Deletion protection can refuse deletion.

The Notion connection check reads the current integration user; Pinecone lists indexes; Supabase checks its REST endpoint. These checks do not create or change service data. Writes in workflow blocks remain writes in the external service. Workspace execution settings and operator provider switches can disable execution.

These native operations do not imply automatic conversion of every imported n8n parameter. Review imported nodes and configure unsupported fields manually before running.

Airtable, HubSpot, Asana and ClickUp workflow nodes

These providers add 30 native operations. Save a personal/private-app token as the indicated credential type, then select that account in the workflow block. A raw token or JSON with accessToken, access_token, token or apiKey is accepted; HubSpot also accepts appToken. These adapters do not add OAuth consent or token renewal.

Lists return one page. Airtable returns Offset; Asana returns next_page.offset; HubSpot returns paging.next.after. Pass the returned value to the matching input for another page. ClickUp Page starts at zero and returns at most 100 tasks per page. Airtable Filter by formula uses Airtable's formula syntax; Fields is a JSON object of field values. Its upsert requires one to three Fields to merge on, subject to Airtable's field-type restrictions. HubSpot search accepts JSON filter groups, property names and one sort; it is a read even though its HTTP method is POST.

Asana and ClickUp task updates require a task ID and at least one field to change. The catalog's input schema describes that choice; an example containing only the ID is not a complete update. Supplying completed: false for Asana is a change and remains valid.

Test connection only reads: Airtable's token identity, Asana's current user, ClickUp's current user, or up to one HubSpot contact. The HubSpot check requires crm.objects.contacts.read; a token limited to another object can fail this check. Passing a check does not prove access to every object, base, project or write operation. Existing deployment switches and workspace execution settings apply to each provider. Imported n8n fields still require manual review, and triggers, attachments and provider-specific batch operations are not included by these nodes.

GitLab, Discord, SendGrid and Brevo workflow nodes

Save the token in Integrations, use Test connection, and select that saved account on the block. These nodes use API tokens; they do not add OAuth renewal or automatically translate imported n8n parameters.

GitLab accepts a project ID or an unescaped namespace/project path and URL-encodes it. Issue iid is the project-local issue number. Lists use Page starting at one and Per page, up to 100. File reads return Base64 content and its encoding; text writes create commits and require an explicit branch and commit message. File updates can include Last commit id to detect concurrent changes. Deleting a release preserves its tag. File writes and releases can trigger your repository's configured automation.

Discord IDs must be strings, preserving their full precision. Message lists use Before; member lists use After. Listing members requires the privileged GUILD_MEMBERS intent; message content visibility and all operations still depend on the bot's intents and channel/server permissions. Sending and editing messages disable parsed mentions, so text such as @everyone does not automatically notify the server. Message bodies are limited to 2,000 characters. Attachments, interactive send-and-wait workflows and event triggers are not included.

SendGrid Get sample contacts returns at most 50 contacts and is not a complete listing. Search contacts uses SendGrid's query syntax. Upsert and selected-contact deletion queue asynchronous jobs; acceptance does not mean processing has finished. Get contact import status checks an upsert/import job. Upserts and deletions accept at most 100 contacts per call here; deletion requires explicit IDs. Lists expose a page token; Delete list preserves its contacts. Email requires a verified sender and sends a single plain-text personalization to at most 100 recipients.

Brevo contact identifiers are an email address or numeric ID. Contact and list reads use a numeric Offset. Contact creation rejects an existing contact; Create or update contact explicitly allows updating one and can return no ID when it updates. Use existing uppercase attribute names. Attribute creation supports normal text/date/float/boolean attributes; option updates target category attributes. Creating a sender does not complete Brevo's sender verification. Template sending requires an existing enabled transactional template and uses its sender/content settings. Both email actions accept at most 99 recipients here.

Connection tests only read: GitLab's current user, Discord's bot user, SendGrid's granted scopes or Brevo's account. A passing check does not prove permission for every resource or write. Existing deployment switches, workspace execution settings and write review apply. The new adapters have local synthetic HTTP and saved-workflow coverage; live service calls from this branch and hosted account setup remain unverified.

Todoist workflow nodes

Save your Todoist API token as a todoistApi credential, use Test connection, and select the saved account in the node. The adapter uses the unified API v1 at api.todoist.com/api/v1; it does not use the retired REST v2 endpoints. The connection test only reads the current user and does not prove access to every shared project or paid feature.

The 39 native operations cover tasks (list/get/create/update/delete/complete/reopen/move/quick add), projects (list/get/create/update/delete/archive/unarchive/collaborators), sections, task or project comments, labels and reminders. Section/comment/label nodes support list/get/create/update/delete. Reminders support list/get/delete and separate relative/absolute create and update operations.

Use string IDs, including numeric-looking legacy IDs. Lists return one page in results and preserve next_cursor; pass that cursor back with the same filters for another page. Limit defaults to 50 and is bounded to 200. Active-task listing can filter by project, section and label name. Completed-task history is outside this adapter.

Task creation requires Content. Task updates require an ID and at least one change: content, description, label names, priority (1–4) or a natural-language Due string. Quick add task uses Todoist's own text parsing. For Move task, choose one destination kind (project, section or parent) and its ID. Comments similarly select one target kind and ID. Project, section and label updates also require at least one changed field; a name is optional. Explicit isFavorite: false is a valid change. The catalog schema explains each choice; an ID-only example is incomplete. These operations do not expose every optional Todoist field.

Relative reminders use Minute offset relative to a task with a due time. Absolute reminders use an ISO 8601 Due date, with an optional timezone. Choose the matching update operation for an existing reminder's type. Notification service is email or push; urgency and availability depend on your Todoist account and devices. Archive/unarchive returns the updated project; successful complete/reopen/delete actions return an empty object. Writes use one attempt and are not automatically replayed after an uncertain failure.

Existing workspace/deployment controls apply through WORKFLOW_LIVE_TODOIST. These nodes do not add OAuth renewal, trigger delivery, file attachments, completed-task history or automatic n8n parameter conversion. Local protocol and saved-workflow tests use synthetic responses; real Todoist calls and hosted account setup remain unverified. The protocol reference is the official Todoist API.

Linear workflow nodes

Connect a personal Linear API key (linearApi), test it, and select the saved account in the node. Personal keys are sent without a Bearer prefix. A saved, currently valid OAuth access-token JSON value (access_token) uses Bearer authorization; this adapter does not add token renewal. Test connection sends a read-only GraphQL viewer query using POST. It checks both HTTP status and the GraphQL response, without creating or updating anything.

The eleven operations get the current user, list teams/users/workflow states/issues, get/create/update/delete an issue, add an issue comment or attach a link. Team, user and state listings help you find the IDs needed by writes. Lists return one page: First defaults to 50 and allows 1–100, and After takes the previous pageInfo.endCursor while hasNextPage is true. Issue listing can filter by team ID. Outputs retain the GraphQL data wrapper and page information.

Issue creation requires a team ID and title. Updates require an issue ID and at least one changed field: title, description, priority, assignee ID or state ID. Priority 0 means no priority and is a valid change. An ID-only catalog example is incomplete; the published input schema describes that choice. The delete operation explicitly disables permanent deletion. Comments require a body; link attachments require an HTTP(S) URL and title. These nodes do not upload files or implement triggers and do not expose every Linear field.

A response with GraphQL errors fails even if HTTP status is 200 and some data is present. A mutation returning success: false also fails. Provider error bodies are not displayed, and uncertain writes are not replayed automatically. Existing team permissions, selected-account access and workspace/deployment controls (WORKFLOW_LIVE_LINEAR) apply. Local protocol and saved-workflow coverage uses synthetic responses; live Linear calls and hosted account setup remain unverified. See the official GraphQL reference.

Trello workflow nodes

Connect Trello with both Trello API key and Trello API token in the named fields. The key is the 32-character key from a Trello Power-Up; the token must authorize the account and required read/write access. The saved trelloApi credential contains apiKey and apiToken. Advanced JSON remains available. A token alone is insufficient. Test connection only reads the current member; it does not create a board or prove write access. Requests send the pair in an authorization header.

The 40 native operations read the current member and workspaces, list/get/create/update/delete boards, list/get/create/update lists, list/get/create/update/delete cards, manage card comments, checklists/items and board labels, and add/remove/list card members. Creating a board requires a workspace ID and creates a private board without default lists or labels. Find that ID with List member workspaces. Deleting a board or card is permanent; use Closed to archive instead. Lists are archived or restored through Update or archive list.

Keep IDs as text; board/card reads also accept their supported short links. Updates require at least one changed field. closed: false restores a board/list/card; dueComplete: false is a valid card change. List id on a card update moves it. Position accepts top or bottom. Checklist items use complete/incomplete, and label color none removes the color. These nodes expose a subset of each resource's fields.

Card and comment reads return one page, defaulting to 50 and bounded to 100. Use Before with the final returned card/action ID for another page; it follows creation order, not visual board position. Member boards and board lists have explicit open/closed filters. Board-label reads return up to the selected limit (at most 100 here); they do not promise every label. Other collection reads remain subject to Trello's API response limits. Writes use one attempt and are not automatically replayed after an uncertain failure.

Workspace/deployment controls apply through WORKFLOW_LIVE_TRELLO. This adapter does not add board invitations/membership administration, attachment uploads, custom fields, webhooks, OAuth consent/renewal or automatic n8n parameter conversion. Local protocol, form and saved-workflow checks use synthetic accounts/responses; live Trello calls and hosted account setup remain unverified. See Trello authentication and API resources and paging.

Freshdesk workflow nodes

Save a freshdeskApi account with its Freshdesk subdomain and Freshdesk API key. For acme.freshdesk.com, enter acme. The account fixes the tenant host; arbitrary URLs are not accepted. Test connection only reads the current agent. Other agent endpoints can require administrator access.

The 28 operations cover tickets, conversations, contacts, companies, agents and groups. Ticket/contact deletion is soft deletion; separate restore operations are available. Company deletion has no restore operation here. Replies and notes accept HTML bodies; notes default to Private. Set it to false deliberately for a public note. Updates require at least one selected change, preserving false and empty tag/domain arrays.

Resource IDs are positive integers. Use Expression to map a numeric ID from an earlier node, or a boolean into Private; the value must have the field's actual type. See parameter mapping.

Lists return one page: Per page defaults to 50, at most 100; Page is bounded to 300. Ticket listing defaults to recent tickets; set Updated since to include older updates. Ticket search allows ten pages and takes the query without outer double quotes.

Workspace/deployment controls use WORKFLOW_LIVE_FRESHDESK. Attachments, triggers, OAuth renewal and automatic n8n conversion are outside this adapter. Local protocol, editor and saved-workflow checks use synthetic responses; live Freshdesk and hosted account setup remain unverified. See the Freshdesk API reference.

Zendesk workflow nodes

Save a zendeskOAuth2Api account using Zendesk subdomain and a currently valid Zendesk OAuth access token, then select it in the node. The account fixes the tenant host. This form does not issue or renew tokens or accept the deprecated API-token/email pair. Test connection only reads the current user and rejects an anonymous identity even after HTTP 200; a passing check does not certify write permissions.

The 35 operations cover tickets/comments/tags, suspended tickets, users, organizations, groups and ticket fields. IDs are positive integers; use Expression to map typed values. Reads preserve Zendesk's response envelope. User outputs omit the browser authenticity token.

Lists return one cursor page, default 50 and at most 100. Pass meta.after_cursor as After while meta.has_more is true. Ticket listing excludes archived tickets. User search uses numbered pages (bounded to 100 here); ticket search adds type:ticket and allows ten pages. Neither search promises every matching record.

Comments use plain text and default to private internal notes (Public false). Updates need a selected change. Ticket Tags replaces the full list; Add ticket tags and Remove ticket tags change only the selected tags (1–100 per call here) and cannot change closed tickets. Create end user fixes the role to end user and skips verification email unless Send verification email is enabled. Updating a user's email adds a secondary identity rather than replacing the primary one. Single suspended-ticket recovery makes the authenticated agent its requester. Review that change before running.

Workspace/deployment controls use WORKFLOW_LIVE_ZENDESK. Attachments, triggers, bulk jobs, OAuth issuance/renewal and automatic n8n conversion are outside this adapter. Local checks use synthetic responses; live Zendesk and hosted setup remain unverified. See authentication, tickets, users and suspended-ticket recovery.

Stripe workflow nodes

Save a stripeApi account using Stripe API key and select that account in the node. Use a secret or restricted API key for the intended test or live Stripe account. Existing raw keys and JSON with secretKey are accepted. A webhook signing secret alone is insufficient. The workflow uses the selected workspace account; it does not borrow the platform billing key. Test connection only reads the balance and needs balance-read permission; it does not certify write access.

The 31 operations cover balance; customers; existing charges and their metadata; coupons; products; prices; payment-method lookup, attachment and detachment; billing meter events; and payment-intent status reads. Requests use the pinned 2026-09-30.endive API version. Payment-intent outputs omit client secrets and next-action URLs. These nodes do not create, confirm or capture payments, collect card details or use deprecated charge creation.

Amounts use integer smallest currency units. A price amount of 0 creates a free price; explicit Active false is preserved. Create recurring prices with day/week/month/year intervals and a positive interval count; Stripe enforces its maximum three-year interval. Existing price amounts cannot be changed here; set Active false to archive a price. Coupon creation has separate percentage and fixed-amount operations and supports once or forever duration.

Metadata accepts up to 50 text key/value pairs, keys up to 40 characters without brackets and values up to 500 characters. An empty value clears one key; an empty object clears all metadata. Updates require at least one selected change. Lists return one page, default 50 and at most 100: use the last ID as Starting after while has_more is true.

Deleting a customer is permanent and immediately cancels their active subscriptions. Detaching a payment method prevents its future reuse or reattachment. Product deletion is allowed only when Stripe's deletion constraints are satisfied, including having no associated prices. Deleting a coupon stops new redemption; existing discounts remain. Review these operations before running them.

A meter event requires an explicit identifier and a nonempty string-valued payload using the keys configured on the existing meter (usually stripe_customer_id and value). Stripe deduplicates identifiers within a rolling window of at least 24 hours. Optional Unix timestamps must be within the past 35 days or at most five minutes ahead. This adapter does not create meters.

Workspace/deployment controls use WORKFLOW_LIVE_STRIPE. Uploads, triggers, OAuth renewal and automatic n8n parameter conversion are outside this adapter. Local checks use synthetic accounts and responses; real Stripe calls and hosted account setup remain unverified. See API versioning, metadata and meter events.

Mailchimp workflow nodes

Save a mailchimpApi account using Mailchimp API token, then select it in the workflow block. Use an API key for your own account or a currently valid OAuth access token; existing mailchimpOAuth2Api credentials are also recognized. Requests use Bearer authentication at the fixed Marketing API origin https://api.mailchimp.com. This form does not issue or renew OAuth grants. Test connection reads account metadata and does not certify campaign-send permissions.

The 23 operations read account information, audiences and interest groups; list/get/create/update/upsert/archive/permanently delete members; list/add/remove member tags; and list/get/delete/replicate/create-resend/send existing campaigns, inspect their content and read the send checklist. These nodes do not create audiences or author new campaign content.

For an existing member, Subscriber hash is the returned member id, or the MD5 hash of their lowercase email. Map {{node.lookup.json.id}} from an earlier member read; do not enter an email address in this field. Create-member uses Email; upsert requires both the hash and email, plus Status if new. Creation requires an explicit Status. Choose it for the person's actual subscription: subscribed skips the double opt-in confirmation, while pending requests confirmation for double opt-in. Check the audience's double_optin value first. Mailchimp's contact guide explains these choices.

Updates require a selected change. False Vip, false interest flags, numeric zero merge fields and empty text values are preserved. Member New email changes the address while the hash identifies the current one. Tags take 1–50 names here; adding a nonexistent active tag creates it. Archive removes a member from the audience; permanent deletion erases their personal data and prevents re-import.

List operations return one page with total_items. Count defaults to 50, maximum 1,000; increase Offset for the next page. Campaign Send sends the existing campaign to its configured recipients; RSS campaigns follow their schedule. Inspect the content, send checklist and recipient settings before running it. A successful send request does not prove delivery. Replicate and create-resend are distinct provider actions; inspect the returned campaign state before any later send. Uncertain writes are not replayed automatically.

Workspace/deployment controls use WORKFLOW_LIVE_MAILCHIMP. The beta Audiences API, SMS, bulk jobs, uploads, triggers, OAuth consent/renewal and automatic n8n conversion are outside this adapter. Local checks use synthetic accounts and responses; real Mailchimp calls, hosted setup and organization-scoped account interaction remain unverified. See authentication and the Marketing API reference.

Google connection permissions

Google OAuth now has separate traditional and agent client profiles. The default Integrations consent uses the traditional profile. A session-only consent started with googlePurpose=agent creates a new personal agent credential; it does not overwrite the existing traditional connection. The administrator must register the agent client independently. Google classification does not replace app verification or the permissions the user grants.

The selected agent Client ID must differ from Google login and both the deployment and workspace traditional Client IDs. Agent custom REST tools renew expired OAuth grants with the saved issuer credentials and save the refreshed value only while the credential is unchanged and accessible. An incomplete imported agent refresh credential requires reconnection; it cannot borrow a traditional client's secret.

Agent REST tools recheck credential access and the saved value immediately before dispatch, including when the token has not expired. Removing, replacing or withdrawing access to the credential stops cached calls. A client ID written into imported JSON is not proof of the token's issuer: native grants carry server-signed issuance metadata, and imported access tokens are verified with Google before use. A token from a traditional client requires new agent consent; a verification outage requires retrying once Google access is restored. Native and verified imported agent grants require a private server signing key in every environment; the deployment administrator must configure it before connecting an account.

For official Google Workspace MCP servers, use Settings → MCP Servers → Google Workspace MCP. Select one service and a permission profile there; these permissions are separate from the existing REST bundle described below. See Google Workspace MCP.

Agent REST destinations must use HTTPS. Service-specific content-* aliases, such as content-gmail.googleapis.com and content-sheets.googleapis.com, require the same agent credentials and scope approval as their regular Workspace API hosts. Multipart batch requests are unavailable for agents; use separate requests. Single-request JSON methods such as Docs batchUpdate remain available. When Google omits permissions from a renewal response, Confluye verifies the renewed token before using it, including for a previously signed grant. Agent REST grants use the registered client's connection access policy. If an administrator explicitly selects verification-only access, the verified scopes must be approved or the caller must match its exact verification audience. The final credential and execution checks run at transport initiation; Google REST custom tools do not follow redirects or A2A status URLs. Official Google MCP hosts cannot be called through custom HTTP tools.

Choose an individual Google service when your workflow needs only that service. The combined Google Workspace connection requests Gmail, Drive, calendar discovery, event management and availability permissions. Its Drive permission also authorizes Docs, Sheets and Slides content operations; separate duplicate content permissions are not requested in that combined consent. Individual Docs, Sheets and Slides connections retain their service-specific permissions.

This reduction removes redundant permission requests. It does not limit the access already provided by full Drive, and it does not revoke a previously granted permission. Disconnect and reconnect if you want to review a fresh consent screen.

Gmail and Trash

Gmail connections request gmail.modify, which supports reading, sending, organizing mail and moving it to Trash. The Google Workspace connection uses the same Gmail permission alongside its other service permissions.

For Send message, use Text or Body for plain text, or HTML for HTML content. If Text and Body are both supplied, non-empty rendered Text takes precedence. If a Text expression renders empty, Body is used.

Use Move message to Trash or Move thread to Trash in a Gmail block. The thread action moves all messages in that conversation. You can restore mail from Trash in Gmail while it remains there. These actions retain their confirmation requirement.

Existing saved workflows and imported n8n Gmail Delete actions now move mail to Trash too. Their operation IDs remain delete-message and delete-thread. Successful outputs include trashed: true; the older deleted: true field remains for compatibility and means the move succeeded.

Existing connections with full Gmail access remain usable. Requesting a smaller scope does not remove an earlier Google grant. To grant only the currently requested permissions, remove the app's existing access in your Google Account, then reconnect the affected services in Integrations. Removing a shared grant can disconnect other services that use it.

Using your own OAuth Client ID and Client Secret still requires Google consent. Google's unverified-app warning depends on the OAuth application's verification status and requested permissions.

The individual Docs connection requests document access only. Calendar requests calendar discovery, event management and availability permissions; it does not request permission to manage sharing or delete calendars. Existing full Calendar grants remain usable. Older event-only connections must reconnect before using Check availability.

Registered Google OAuth clients allow all authorized users to start consent by default. This availability setting does not claim that Google has approved the application or its permissions; Google still controls its consent screen and any provider restrictions. Workspace permissions, MFA and callback checks still apply.

Administrators can explicitly set connectionAccess: "verification-only" to restrict a client to reviewed permissions or an exact configured verification audience. A 403 with google-oauth-verification-required means the client is unregistered, its release configuration is invalid, or that optional restriction refused the request. The deployment administrator must correct the client registration or access policy; repeatedly reconnecting will not clear it. Restricted access is checked again when Google redirects back. Saved connections continue to use their recorded grants.

Copy and move Drive files; replace text in Docs

Select a Google Drive or Google Workspace connection for Copy file and Move file. Both use fileId for the source file. Set parentFolderId to the destination folder; it is required for Move file and optional for Copy file. Copy file also accepts an optional name for the new file. Open the result in Drive to check its name and location.

For Google Docs Replace all text, select a Docs or Google Workspace connection, enter documentId, and add replacements entries with search and replace. An empty replace removes the matching text. For example, [{"search":"BEFORE","replace":"AFTER"}] replaces every occurrence of BEFORE. Open the document to verify the change. These actions use the existing Drive and Docs permissions.

Imported workflows can also use the older folderId, fileName, text, and replaceText fields. Non-blank older destination, name, and search values take precedence; if they render as empty or whitespace, the fields above are used. An empty replaceText still removes matching text, even when replace is also present.

Create a Google Sheets tab

In the workflow builder, add Google Sheets: Create sheet to create a tab inside an existing spreadsheet.

1. Select your Google Sheets or Google Workspace connection.
2. Choose the existing spreadsheet or enter its spreadsheetId from its Google Sheets URL.
3. Set title to the new tab name, then test the block.
4. Open the returned url to check the new tab. Use the returned sheetId in later blocks that need a tab identifier; use the tab title in A1 ranges for reading or writing cells.

The action returns spreadsheetId, sheetId, title, index, and url. Create spreadsheet creates a separate file; Create sheet adds a tab to the selected file. This uses the existing spreadsheets permission and requires no additional OAuth permission.

A duplicate tab name produces a Google error; choose another name. The action does not retry automatically. If a request times out, check the spreadsheet before running it again, because Google may already have created the tab.

WhatsApp Business Cloud

Add WhatsApp: Send message or WhatsApp: Send template in the workflow builder’s Communication catalog.

1. Create or select a WhatsApp account. Enter its Access Token and Business Account ID from Meta’s WhatsApp → API Setup in separate fields. The Business Account ID is the WABA ID, not the Phone Number ID. The token needs whatsapp_business_messaging permission for sending. The credential type is whatsAppApi; existing token-only and JSON credentials remain usable for their previous sending flows.
2. Save and test checks token validity without sending anything. It does not prove that the token can send from a particular phone number.
3. Set phoneNumberId to the sender’s Phone Number ID from Meta, not its telephone number or Business Account ID. Set recipientPhoneNumber to the recipient’s international number, including country code, for example +14155552671.
4. Send message accepts text (up to 4,096 characters) and optional previewUrl. Free-form messages require an open customer service window, normally 24 hours after the customer’s last message. Outside that window, use Send template.
5. Send template accepts the exact approved templateName, its languageCode (for example es_MX), and optional ordered bodyParameters entries such as [{"text":"Ana"},{"text":"123"}]. This operation supports text body parameters; headers, buttons, media and interactive messages are not supported.
6. Test with an authorized recipient. Meta test numbers require recipients configured in the Meta developer console. The output contains messageId and status: "accepted"; acceptance is not a delivery receipt.

The connector makes one attempt per execution. Leave the node’s Retry On Fail setting off to avoid duplicate messages. If a send times out or returns no receipt, check Meta before running it again. For rejected requests, check token expiry, sender access, recipient setup and template approval. Sending does not track delivery updates. Operators can disable WhatsApp sending with WORKFLOW_LIVE_WHATSAPP=false.

Receive WhatsApp messages

Add a WhatsApp trigger. From the trigger, choose Add account to open the WhatsApp credential form; its Access Token and Business Account ID fields are both visible and required. Enter the WABA ID from Meta’s WhatsApp → API Setup, not the sender’s Phone Number ID. Existing token-only accounts can be updated by re-entering the token and adding the WABA ID. The token needs whatsapp_business_management permission. The Meta app must already have its Webhooks product and messages field enabled. Publish the workflow to let Confluye subscribe the WABA with a callback override. Meta may reject registration if the app has not completed its initial webhook setup; the publication error names that prerequisite. One WABA callback override receives the account's messages, so use one trigger per WABA. Confluye answers Meta's verification challenge automatically. Optionally enter the App Secret to verify the signature of incoming POST requests; re-enter it when updating that account.

Send Telegram messages

Ordinary Telegram workflow actions use the explicit public Bot API profile and exact-request approval described in Telegram ordinary actions. They return a bounded synchronous response; later delivery remains unverified. Repair and reselect the account for imported nodes rather than applying old queued-message instructions to the new profile.

Existing alert nodes and chat-bound notifications retain their separate queued-delivery lifecycle. A queued reference confirms acceptance into that queue, not Telegram delivery. Their old account formats and response-wait behavior do not certify the new profile. Incoming triggers use the separate publishing flow below.

Receive Telegram messages

Add a Telegram trigger and select a workspace Telegram account containing the Bot Token, either as a plain token or {"botToken":"…"}. Under Trigger on, choose Messages, Callback queries, or Inline queries; new and existing triggers default to Messages. Publishing calls Telegram setWebhook with the selected update types. The trigger validates Telegram's secret header and starts the workflow when a selected update arrives. Callback query payloads are available as {{input.callback_query}}; inline query payloads as {{input.inline_query}}. A Telegram bot has one active webhook, so publishing another workflow for the same bot replaces its previous callback. A destination chat ID is needed only for sending Telegram messages, not for this trigger.

Existing Telegram and WhatsApp triggers that use a manually registered webhook keep their current URL and continue to work without editing. In one of those nodes, select a workspace account to switch it to automatic registration when you next publish.

After changing a selected trigger account's token, publish the workflow again so the provider receives the current callback verification token.

Publishing Production or a named preview, including switching a stopped deployment back On, registers the managed callback and its secret header. If registration fails, Confluye reports the provider registration error even though the deployment was saved. Retry publication to register the callback again; a retry of the same publication request keeps the existing activation. Avoid manually calling setWebhook for a managed trigger, because a callback registered without Confluye's secret will be rejected with HTTP 403. Use a separate bot for each simultaneously active destination: publishing a preview with the same bot replaces Production's Telegram callback.

The implementation follows the official Cloud API approach used by n8n’s WhatsApp node. Existing encrypted workspace credentials and workflow connection bindings apply.

PocketBase

Choose PocketBase from Integrations to save a pocketbaseApi credential. Enter the HTTPS server URL without /api, the auth collection (_superusers by default, or a custom auth collection), and the account identity and password. Use an account with access only to the collections and records the workflow needs. The connection test signs in and checks the returned token; it does not read or change records. The password is stored in the encrypted workspace credential; each run obtains a short-lived token and renews it when it expires or PocketBase returns 401.

PocketBase workflow blocks can list records, get one record, create, update, upsert, delete, and upload files to an existing record. List accepts PocketBase filter, sort, expand, and fields expressions and paginates by default; Return all is capped at 10,000 records. Upsert finds the first record matching its required filter, then updates it or creates a record if none matched; concurrent writers can race between those requests. Create, update, upsert, delete, and file upload have one attempt by default because a write may succeed before a timeout. Review node retry settings before enabling retries.

File upload takes a ready Workspace File reference and a target PocketBase file-field name; it patches the existing record and can include other record fields. Collection suggestions require a superuser account. If the account cannot list collections, type the collection name or ID in the field; record access still follows the PocketBase rules for that account. See PocketBase's API records documentation for its filter and record APIs.

File uploads require PocketBase 0.21 or later to preserve JSON record fields alongside the uploaded bytes. PocketBase introduced that multipart format in its 0.21 release; the sign-in-only connection test does not check server-version compatibility.

Jenkins

Add a Jenkins block from the workflow builder’s Developer Tools catalog.

1. Create an API token in Jenkins under your user’s Security settings. Use a dedicated user with only the job permissions the workflow needs. Do not use the account password.

2. Create or select a Jenkins connection. Save JSON with the HTTPS Jenkins URL, including any context path, the user and the token: {"baseUrl":"https://jenkins.example.com","username":"ci-bot","apiToken":"…"}. The credential type is jenkinsApi, and the apiKey field name from n8n is also accepted.

3. Save and test signs in as that user without starting a build. It does not prove that the user can build a particular job.

4. Set jobName to the job name. For a job inside folders, separate the folders with /, for example platform/api/deploy.

5. Trigger build starts a job without parameters. Trigger build with parameters accepts optional parameters entries such as [{"name":"BRANCH","value":"main"}]. Parameters you omit use the job’s defaults. Both return status: "queued" and, when Jenkins reports it, a queueId. Trigger build, and Trigger build and wait for result without parameters, call Jenkins’ /build endpoint. When Jenkins answers 400 because the job declares parameters, they retry once with /buildWithParameters and no values, so the job’s defaults apply.

6. Get queue item uses that queueId. It returns state (waiting, started or cancelled) and, after the build starts, buildNumber and buildUrl. Jenkins only keeps finished queue items for a few minutes.

7. Get build accepts a buildNumber or lastBuild, lastCompletedBuild, lastSuccessfulBuild or lastFailedBuild. It defaults to lastBuild. The output includes building, result (for example SUCCESS or FAILURE, or null while it runs), durationMs, startedAt (null when Jenkins reports no start time) and url. List builds returns up to limit recent builds, newest first.

8. Get build details accepts the same buildNumber values as Get build and returns the same build fields, plus:

   • status: running, success, unstable, build_failure, aborted or not_built.
   • stages: each Pipeline stage with name, status (for example SUCCESS, FAILED or NOT_EXECUTED), durationMs and, when Jenkins reports one, errorMessage.
   • failedStage: the name and errorMessage of the first stage with status FAILED, or null when no stage failed.
   • stagesAvailable: false when Jenkins has no stage data for the build, as with freestyle jobs or servers without the Pipeline Stage View plugin. stages is then empty.
   • logTail: the last logLines lines of the console log (default 200, up to 1,000), without Jenkins console annotations and terminal color codes. At most the last 256 KB of the log is read.
   • logTruncated: true when logTail does not include the whole log.

   To analyze a failure in the same step, turn on includeDetails in Trigger build and wait for result instead. Otherwise, connect Get build details to its build_failure branch, with buildNumber set to that node’s buildNumber. Then pass failedStage and logTail to an agent or AI block to explain the failure. The console log can contain anything the job prints, so the workflow’s later steps receive it as-is.

A trigger makes one attempt per execution. Leave the node’s Retry On Fail setting off to avoid starting duplicate builds. If a trigger times out, check Jenkins before running it again. Once Jenkins accepts the build, a later failure of the node is never retried, even with Retry On Fail on. Reading operations retry once.

Start a workflow when a build finishes

The Jenkins build finished trigger starts the workflow when a build finishes on jobs you name. Jenkins does not call Confluye: the destination polls Jenkins with the connection you choose, so no Jenkins plugin or webhook is needed.

1. Add Jenkins build finished from the Triggers catalog and save the workflow.
2. In Versions & previews, open the destination and choose Add event source. Select Jenkins finished builds, the saved Jenkins trigger and a Jenkins connection. You need permission to edit that destination.
3. Add one row per job. Use the full job name, with folders separated by /. Jobs are never discovered from folders or the server.
4. Choose the Build results that start the workflow. All five are selected by default: SUCCESS, UNSTABLE, FAILURE, ABORTED and NOT_BUILT.
5. Set the poll interval in seconds. The default and minimum are 60 seconds. Then choose Configure source.

A source for a trigger that the destination's deployed version does not have yet shows Off and starts polling when you deploy that version, so builds never start the older version. Changes to a source that is already polling apply from its next poll. Configure source is refused when the saved or deployed version launches one of the listed jobs.

The first poll of each job records the latest existing build, finished or still running, as its baseline. That build and older builds never start the workflow; only builds numbered above the baseline do. Adding a job later takes a baseline for that job only. If the connection starts pointing at another Jenkins server, every job takes a new baseline. A job that is deleted and recreated in Jenkins restarts its build numbers, so it also takes a new baseline. Stopping the destination turns its Jenkins sources off. When you deploy or resume it again, each job takes a new baseline, so builds that finished while it was stopped never start the workflow.

Each finished build starts at most one run per destination, even with retries, several workers or restarts. A build that finishes out of order, for example #14 before #13, still starts its own run. Each poll delivers up to 25 builds per job, oldest first. When more than 100 newer builds arrive before an earlier build finishes, that build is skipped and counted.

The run input contains job, buildNumber, result, durationMs, startedAt (null when Jenkins reports no start time) and url, plus trigger: "jenkins" and the triggerId.

The destination shows the source's state, interval and last poll time. For each job it shows Waiting for first poll or Baseline: build #N, then Last build seen: #N, the number of builds skipped and the latest error. A job Jenkins cannot find (404) shows its own error while the other jobs keep polling. Authentication failures (401 or 403), a blocked host and a deleted connection stop the whole source, show the error and retry with increasing delays.

Set up Jenkins from an MCP client

An AI client connected to the Confluye MCP server can do the same setup. It needs a workspace API key with write and execute authority.

1. catalog.search with jenkins returns the trigger and the Jenkins blocks, whose credential type is jenkinsApi.
2. credentials.create with providerId: "jenkins" stores the connection JSON shown above. credentials.list with providerId: "jenkins" lists existing connections as metadata, and credentials.test runs the same connection test as Settings. No tool returns the API token.
3. workflows.update-target with the configure-source action, the destination's expectedRevision from workflows.targets, and a jenkins source (targetId, the trigger's nodeId, credentialId, jobs, optional results and intervalSeconds) configures the source with the same limits and loop check as Configure source, as the key's user.
4. workflows.targets reports each source's state, interval, last poll and per-job status.

Only a workspace API key can create a credential through MCP. OAuth clients such as ChatGPT can list and test connections and configure sources, but create the connection in Settings → Credentials.

Trigger a build and wait for its result

Trigger build and wait for result queues a build and pauses the run until that exact build finishes or the timeout passes. It never follows another build of the same job. While paused, the run holds no worker and uses no compute credits.

• jobName is required. parameters is optional, in the same format as Trigger build with parameters. timeoutMinutes is between 1 and 10,080 (7 days) and defaults to 60.
• Connect the node’s outgoing edges with the branch labels success, unstable, build_failure, aborted, not_built, cancelled and timeout. cancelled means the queue item was cancelled in Jenkins. A status without its own edge follows the unlabeled edge. Errors from the node itself, such as a rejected request, follow the error branch, never build_failure.
• The output includes status, jobName, queueId, buildNumber, url, timedOut and, when Jenkins reports them, queueWhy and reason. Parameter values are not repeated in the node output.
• Turn on includeDetails to also get stagesAvailable, stages, failedStage, logTail and logTruncated when the build finishes, as described for Get build details. logLines sets how many final log lines to keep (default 200, up to 1,000). They are read when the wait resumes with success, unstable, build_failure, aborted or not_built, never with cancelled or timeout. If they cannot be read, the output includes detailsError instead, and the run still continues on the branch for the build result.
• After the deadline, a build keeps its real result only if Jenkins’ completion time (its start time plus duration) is at or before the deadline, even if Confluye sees it slightly later. If the timeout passes first, or Jenkins does not report when the build finished, the run continues on timeout and the output includes the build URL once the build has started. A queue item cancellation first seen after the deadline also continues on timeout.
• A timeout or a cancelled run does not stop the build or remove it from the Jenkins queue.
• While the run waits, the node inspector and the run query show the job, queue item, build number once it starts, and the deadline. They do not show the Jenkins server address.
• If Jenkins does not return a queue item, the node fails without pausing, and Retry On Fail does not queue the build again. Check Jenkins before retrying.

This operation cannot run inside a loop or parallel group; deployment is blocked until it moves to the main path. A workflow with a Jenkins trigger also cannot deploy if it launches a job its own trigger listens to on the same Jenkins server, or a job whose name is an expression. A job with the same name on another server is allowed only when the destination can read both connections; otherwise it counts as the same job.

Network access and operator switch

The Jenkins server must be reachable over HTTPS from Confluye. The egress guard blocks private and internal addresses unless an operator adds the host to FLUXUS_HTTP_EGRESS_ALLOWLIST. Every Jenkins request, including each redirect, stays on the exact scheme, host and port of the connection URL; a redirect to another host, subdomain, port or to http:// is refused before anything is sent.

Operators set WORKFLOW_LIVE_JENKINS on both the web service and the worker. WORKFLOW_LIVE_JENKINS=false disables the actions, including trigger and wait, and pauses polling and wait tracking. Waiting runs keep their deadline and continue on timeout when it passes. Builds that finish while polling is paused are not delivered: when it is turned back on, each job takes a new baseline and its status counts the skipped builds. A pause does not count as a polling failure, so it does not lengthen the retry delay after polling resumes.

The operations follow the Jenkins remote access API used by n8n’s Jenkins node. Creating, copying and deleting jobs and restarting or pausing Jenkins are not supported.

Twilio workflow nodes

Connect a twilioApi account using the named Twilio Account SID and Twilio Auth Token fields. Twilio authentication type defaults to authToken. To use an API key instead, set it to apiKey, enter both Twilio API Key SID and Twilio API Key Secret, and leave Auth Token empty. The Account SID always identifies the target account; an SK-prefixed key SID never replaces it in request paths. Credentials stay out of node inputs. This adapter uses Twilio's global api.twilio.com endpoint; other regional endpoints are not configured here.

The 18 operations list/get/send/redact/cancel/delete messages; list/get/create/redirect/end/delete calls; list/get/delete recording metadata; list/get owned phone numbers; and read the balance. Test connection only lists one message and therefore requires message-read permission. A successful check does not establish permission to send messages, call, delete records or read the balance. API keys and Auth Tokens follow Twilio's authentication rules.

Send SMS, MMS or WhatsApp message requires To, a sender (From or Messaging service sid) and content (Body, Media urls or Content sid). Phone recipients use E.164 (+ and country code); WhatsApp addresses use whatsapp:+…. Body accepts up to 1,600 characters and media accepts at most ten HTTP(S) URLs, sent as repeated form parameters. Content templates use an HX-prefixed SID and optional JSON Content variables; configure approval and WhatsApp availability in Twilio. Template content and body/media are separate alternatives. Twilio fetches supplied media/callback URLs; this node does not upload a local file. Explicit smartEncoded: false is preserved.

Calls require E.164 To/From and exactly one of Twiml (up to 4,000 characters) or an HTTP(S) Url serving TwiML. Recording is optional and requires any applicable consent. Redirect active call changes an existing call's instructions; Cancel or end call requests canceled or completed. Deleting a call/message record does not undo a completed call or delivered message. Redaction sends an empty message body. Cancellation applies to a scheduled message that Twilio still allows canceling. Sends and calls can incur charges.

List operations return one page; Page size defaults to 50 and allows 1–1,000. Use the provider's next-page token as Page token; the adapter does not automatically follow next_page_uri. Recording operations return JSON metadata, not audio bytes. Number reads do not purchase or release numbers. Existing execution controls use WORKFLOW_LIVE_TWILIO. Uncertain writes are not replayed automatically. Triggers, webhook receipt, OAuth issuance and regional endpoints are outside this adapter. Protocol, local-control and saved-workflow tests use simulated responses; real Twilio calls and hosted account/organization setup remain unverified. See Messages and Calls.

Pipedrive workflow nodes

Connect a personal Pipedrive API token (pipedriveApi) for the intended user and company. Requests use the fixed https://api.pipedrive.com origin and the x-api-token header, following the current authentication documentation. The token is never appended to a URL. A currently valid manual OAuth secret with access_token uses Bearer authorization; token issuance and renewal are not included. Test connection reads /v1/users/me and checks both success: true and the returned user ID.

The 36 operations read the current user and list/get/create/update/delete deals, persons, organizations, activities, products, pipelines and stages. Those seven resources use API v2, including PATCH for updates. IDs are positive integers. Creation requires a title for deals, a name for persons/organizations/products/pipelines, subject and type for activities, and name plus pipeline ID for stages. Updates require an ID and at least one changed field; false, 0, empty text and selected empty arrays remain intentional changes. Product prices use currency-unit numbers rather than Stripe's smallest-unit integer convention. Custom fields use their 40-character hash as keys; null clears a value, subject to Pipedrive's field rules.

Use Participants for activity contacts, marking one as primary to set the activity's person. Date fields use ISO dates; time/duration inputs use HH:MM. Done and Busy accept false. Organization, person, deal and activity list operations expose their documented ownership/link filters; each list returns one page with Limit (default 50, maximum 500) and Cursor from additional_data.next_cursor. Returned data retains the success/data envelope and pagination fields. Responses with success: false fail even with HTTP 200.

Deletes mark resources deleted; Pipedrive documents permanent removal of deals, persons, organizations and products after 30 days. Pipeline/stage changes require the user's applicable permissions. This adapter exposes a subset of resource fields and does not implement files, notes, leads, triggers, automatic n8n parameter conversion or OAuth renewal. Execution uses WORKFLOW_LIVE_PIPEDRIVE; uncertain writes are not automatically retried. Local protocol and saved-flow tests are simulated; real Pipedrive calls and hosted account/organization setup remain unverified. See the API reference and v2 guidance.

WordPress workflow nodes

Connect wordpressApi using WordPress site URL, WordPress username and WordPress application password. Use the HTTPS site root, including an installation subfolder such as https://example.com/blog; leave off /wp-json, query parameters and embedded credentials. Create the application password in the owning WordPress user's profile, following WordPress authentication guidance. The adapter sends UTF-8 Basic authentication only to the selected site's origin and preserves its installation path. It requires verified HTTPS; it does not disable certificate validation. This is the site's core REST API, not the separate WordPress.com public API.

The 29 operations get the current user; list/get/create/update/delete posts, pages, categories, tags and comments; and list/get/update public user-profile fields. Test connection reads the authenticated /users/me?context=edit profile and requires a positive user ID. Public post access alone does not pass this check. Write permissions and moderation/publishing rules remain those of the selected WordPress user and site.

Post/page creation requires a title and defaults to Draft. Choose Publish, Pending, Private or Future explicitly when needed; supply a suitable date for scheduled publication. Updates require an ID and at least one changed field. Post categories/tags are arrays of term IDs. sticky: false, parent/featured-media ID 0, menu order 0 and selected empty arrays remain valid changes. Feature/media ID zero removes an assigned image. Optional blank text fields are omitted by the form; to clear content, map an expression such as {{input.content}} that resolves to an empty string. The saved-workflow regression covers that case. Returned post/page protection passwords are omitted, including inside permanent-deletion results.

Post, page and comment deletion defaults Force to false, requesting Trash. Choose true only for permanent deletion; availability of Trash depends on the site's configuration. Category/tag deletion is always permanent and is labeled accordingly. User-profile updates cover name, URL, description, nickname and slug; this adapter does not create/delete users or change roles, login passwords or application passwords.

Collections return one JSON array page. Per page defaults to 50 and allows 1–100; Page starts at 1. Search is available on every collection; resource-specific filters cover status, author, terms, parent, hide-empty and comment post/status where shown. HTTP pagination headers are not included in the node output and pages are not fetched automatically. A page beyond the available range can fail in WordPress. These operations expose a subset of core fields; media uploads, custom post types, plugin endpoints, OAuth, triggers and automatic n8n parameter conversion are not included. Existing execution controls use WORKFLOW_LIVE_WORDPRESS, with one attempt per uncertain write. Local protocol, controls and saved-flow tests are simulated; real WordPress requests and hosted account/organization setup remain unverified. See the REST reference.

WooCommerce workflow nodes

Connect wooCommerceApi with WooCommerce store URL, WooCommerce consumer key and WooCommerce consumer secret. Use the HTTPS store root, including an installation subfolder when needed, without /wp-json or query parameters. Generate the ck*/cs* pair under WooCommerce → Settings → Advanced → REST API for the intended store user and permissions. Requests use Basic authentication in a header, following the current authentication guide. Query-string credentials and key generation/renewal are not implemented. These are administrative REST v3 nodes, separate from the public Store/Checkout API.

The 30 operations list/get/create/update/delete products, orders, customers, coupons and product categories/tags. Test connection reads at most one product category, requiring key read permission. It checks that the response is a category array; an HTML login/frontend page or invalid JSON object does not pass. This does not certify write permissions. Selected credentials keep requests bound to that store and installation path.

Product creation requires a name and defaults to Publish, matching WooCommerce; choose Draft, Pending or Private when appropriate. Orders require an explicit Line items array, which can be empty, and default to Pending with Set paid false. True marks an order paid, changes status and reduces stock; it does not collect a gateway payment. Store actions can trigger the site's configured notifications and automations. Customers require an email; account generation and notifications follow store rules. This port does not accept customer login passwords or change customer roles.

Prices and discount amounts use currency-unit decimal strings, such as "0" or "19.95", not Stripe smallest-unit integers. Updates require an ID and a selected changed field. Requests preserve false flags, numeric zero, empty selected arrays and JSON metadata values without renaming their custom keys. Address and line-item fields use the named camel-case controls; the adapter sends WooCommerce's snake-case property names. For order lines, supply Product id to add a line or Id to edit an existing line; quantity zero removes an existing line. Image inputs use an existing image ID or an HTTP(S) source URL fetched by the store. To clear optional text/price fields, map an expression that resolves to an empty string; blank optional literal controls are omitted. Returned login passwords, order_key and payment_url are omitted; this adapter does not expose checkout access links.

Product/order/coupon deletion defaults Force to false, requesting Trash. True requests permanent removal. Customer/category/tag deletion is always permanent and is labeled accordingly. Each list returns one JSON array page; Per page defaults to 50, maximum 100, and Page starts at 1. Search and the resource-specific filters shown in the node are supported; order Customer id becomes the API's customer list filter. Response pagination headers and automatic following of Link URLs are not included.

These nodes expose a subset of REST v3 fields. Variations, refunds, taxes/shipping methods, batch jobs, webhooks, OAuth and automatic n8n parameter conversion are outside this port. Execution uses WORKFLOW_LIVE_WOOCOMMERCE and existing workspace/account controls; uncertain writes are not automatically replayed. Local protocol, real-control and saved-flow tests use simulated HTTP. Real WooCommerce calls and hosted account/organization setup remain unverified. See the REST v3 reference.

Baserow workflow nodes

Connect baserowApi using Baserow host and Baserow database token. Use an HTTPS API origin without a path, query or embedded credentials; Cloud defaults to https://api.baserow.io. Create a database token with the intended tables and create/read/update/delete permissions, following the database token guide. This form does not accept the username/password login used by the older n8n reference node. Tokens use the Authorization: Token header and requests stay bound to the selected host. Key creation, login and renewal are not included.

The 11 operations list accessible tables and table fields; list, get, create, update, move and delete rows; and create, update or delete rows in a batch. Test connection reads the accessible tables endpoint and rejects an invalid table response. A successful check does not certify access to a particular table or permission to write. Use the list operation to find numeric table IDs, and list fields or the table's generated API docs to check writable field formats.

User field names defaults to true: enter exact column names in the Fields JSON object. Set it false to use keys such as field_123. Values preserve text, empty strings, false, zero, null, objects and arrays. Formula/read-only fields and table-specific formats remain subject to Baserow validation. Create requires an explicit Fields object; {} deliberately creates a row using store defaults. Update requires at least one field. Map a whole Fields expression to an object; it is not a JSON string. Link-field updates replace their selected relations, and an empty array clears them.

Row lists return one page with count, next, previous and results. Size defaults to 100 and is limited to 200; Page starts at 1. Search, order, JSON filters, view ID and include/exclude are supported. Pagination URLs are returned as data and are never automatically followed. For moving a row, Before id selects the row it should precede; leaving it blank moves it to the end. Create uses the same optional Before id to position new rows.

Batches accept 1–200 items. Create/update use an Items array of row objects; every update item needs its numeric id and at least one column change. Batch deletion takes Row ids. Deletion returns an empty success object and follows the server's row-deletion behavior; recovery is not exposed here. Batch availability, permissions and webhook behavior depend on the server/version; the current API documents a limitation for deleted-row webhooks in batch deletion. The nodes do not suppress normal server notifications or webhooks. File uploads, schema changes, triggers and automatic n8n parameter conversion are outside this port.

Execution uses WORKFLOW_LIVE_BASEROW and existing workspace/account controls. Uncertain writes are not automatically replayed. Local protocol, actual-control and saved-flow checks use simulated HTTP; real Baserow calls and hosted Review/organization setup remain unverified. See the database API reference and OpenAPI specification.

Contentful workflow nodes

Connect contentfulApi with Contentful space ID, Contentful access token, Contentful API source and Contentful API region. Source defaults to delivery for published content; preview reads the latest published and draft content and needs its separate preview token. Region defaults to us; choose eu for an EU data residency space. Requests use Bearer authentication in a header and stay bound to that account's space and official delivery/preview origin. Token creation, OAuth and renewal are not included. See delivery authentication and preview authentication.

The ten read operations get the selected space; list/get content types, entries, assets and tags; and list locales. Test connection reads only the selected space and rejects a different space or malformed successful response. It does not certify access to each environment or content item. Environment id defaults to master; the token must permit the selected environment. Resource IDs are text, including numeric-looking IDs, and each account remains tied to its saved space.

Entry and asset reads support Locale; * returns values grouped by locale, so mapping paths differ from a single-locale result. Collection controls expose bounded offset pagination: Limit defaults to 100, maximum 1000, and Skip starts at zero. Locales use the server's collection response without extra pagination controls. These nodes return the response envelope, including items, counts, linked includes and unresolved-link errors; they do not flatten items or fetch further pages. Offset pagination is for bounded reads; cursor traversal and synchronization are outside this port.

Entry/asset lists support full-text Query, Order, Select and scalar Filters JSON keys starting with fields., sys. or metadata.. For example, {"fields.count":0,"fields.active":false} preserves both scalar values when sent as query parameters. Entry field filters require Content type id. Server rules still determine supported operators and fields. Entry collections expose Include depth 0–10, default 1; zero omits linked items. Assets optionally filter by MIME type group. Asset reads return file metadata/URLs and do not download or upload files. See search parameters and linked content.

These nodes do not create, edit, publish or delete content, issue asset keys, or add GraphQL, taxonomy, triggers or automatic n8n parameter conversion. Execution uses WORKFLOW_LIVE_CONTENTFUL and existing workspace/account controls. Local protocol, real-control and saved-flow checks use simulated HTTP. Real Contentful delivery/preview/EU calls, environment permissions and hosted Review/organization setup remain unverified.

MailerLite workflow nodes

Connect mailerLiteApi using MailerLite API token for the current MailerLite API. Requests use Bearer authentication at https://connect.mailerlite.com; Classic API keys and classicApi:true are rejected. Enter the current token in the named field, which saves apiKey JSON; Advanced JSON also requires valid JSON. This port does not issue or renew tokens. Test connection reads one group page and rejects invalid group data; success does not certify subscriber writes, imports or access to ecommerce shops. See authentication.

The 23 operations cover subscriber list, ID/email lookup, upsert, update, removal, scheduled forgetting, total count, activity, import progress and ecommerce statistics/orders; group list/create/rename/delete, membership changes, group subscribers and bulk import; and custom-field list/create/rename/delete. All IDs are text, including large decimal IDs such as "31986843064993537", to avoid numeric rounding. Subscriber lookup also accepts an email address.

Upsert requires an email and adds supplied fields/groups without removing omitted values. Resubscribe true explicitly requests re-subscription; the provider's status rules still apply. Update requires at least one selected field and uses PUT. Enter Groups as a JSON array of text IDs, for example ["31985378335392975"]. Supplying Groups on update replaces membership; [] removes all groups. Fields is JSON with exact default/custom field keys and text, numeric or null values. Empty text, zero and null remain explicit; fields are not removed by omission. Timestamps use yyyy-MM-dd HH:mm:ss, and the named IP fields accept IPv4/IPv6. Subscriber and group changes can trigger configured automations; these operations do not directly send campaigns.

Subscriber lists use an opaque Cursor, default Limit 25; group-member lists default to 50 and the server defaults their omitted Status to active. Include groups optionally returns membership. Groups, fields and activity use page numbers starting at 1, default Limit 100. Ecommerce orders default to 10 and need a Shop id. Native page limits are 1–1000. Every list returns its complete data/links/meta envelope as supplied; pagination URLs are never automatically followed. Total subscriber count has a separate operation.

Bulk group import takes an explicit nonempty Subscribers array of email/optional Fields objects and returns a progress URL. Acceptance does not mean the import is complete; copy the numeric import ID from the end of that URL into Read import progress. The node does not follow the URL or poll automatically. The current API limits import creation to five requests per minute per token. Normal subscriber removal keeps information at the provider in case of re-subscription; Schedule permanent subscriber data removal requests the provider's separate forget operation, documented to complete within 30 days. Group/field deletion follows provider behavior; restoration is outside this port. See subscribers, groups and fields.

Campaign sending, form/segment/automation management, triggers, Classic API support and automatic n8n parameter conversion remain outside this port. Execution uses WORKFLOW_LIVE_MAILERLITE and existing workspace/account controls; uncertain writes are not automatically replayed. Local protocol, actual-control and saved-flow checks use simulated HTTP. Real subscriber/group changes, automations, scheduled forgetting, import completion, shop permissions and hosted Review/organization setup remain unverified.

Credentials and secrets

For workflow Versions & previews, choose compatible connections explicitly for each destination. Each field filters by provider and required permissions. Add credential opens the existing provider dialog without leaving the deployment setup; save and test a workspace credential, then save the destination’s connection choices. OAuth sign-in returns to the same destination, where you can select the connected account. Production and a preview can share an AI connection while using different CRM accounts. A shared account still reaches the same real data; a separate preview URL does not isolate that account. AIMS approval covers the exact destination and connection references. Rotated secret values are resolved when executing, while missing, inaccessible or retired connections block execution. Creating a preview does not copy provider subscriptions, mailbox cursors or pinned test data. External secret-provider references from the former environment catalog are not automatically converted into workflow connections.

Values are encrypted with AES-256-GCM (APP_ENCRYPTION_KEY, or a dev-only default). List/create/update responses expose maskedValue (•••••••• from Prisma mapping). OAuth scope strings may be parsed from JSON secrets into oauthScopes.

Create requires name and type. Optional envVar is not an arbitrary environment variable: it must be a known deployment source (DATABASE_URL, GITHUB_TOKEN, OPENAI_API_KEY, SMTP/Google/Jira/Microsoft/Supabase JSON, etc.) whose type matches the credential. Otherwise the API returns a public error and does not store the secret.

Claude Code / Codex CLI types (claudeCodeOAuth, codexCliSession, or those display names) cannot be created or patched through credentials REST/tRPC. Use the local AI CLI worker login on Settings → Workers.

POST /api/v1/credentials/{id}/test decrypts the secret and runs testCredentialConnection. Unlike list/create/update/delete, the test route does not require workspace-scoped keys — any valid bearer key for that workspace id can test.

Personal credentials: tRPC createCredential accepts scope: "personal" and binds userId. The v1 REST create path always stores scope: "workspace".

Authorization

There is no Admin/Owner check on credential CRUD. Cross-workspace ids return 403/404.

Failure modes

• Invalid type/value format: 400 from validateCredentialSecretValue.
• Duplicate name for the same scope: create updates the existing row instead of erroring.
• Decrypt failure at test/runtime: SecretDecryptionError — reconnect the credential.
• OAuth start without key: 400. Foreign workspaceId: 403.

Next steps

Intercom workflow nodes

Connect intercomApi using Intercom access token and Intercom API region (us, eu or au, default us). Save a raw access token without Bearer; a raw token credential uses the US region. Requests stay on the selected official regional origin and send Intercom-Version: 2.16. Test connection reads the current API admin from /me; it does not certify contact writes or all permissions. These nodes do not issue or renew tokens. See Intercom authentication and API reference.

The 38 operations cover current admin, contact list/search/get/create/update/delete/archive/unarchive/merge, companies, contact/company links, notes, tags, segments and contact subscription preferences. Contacts use current /contacts routes, rather than Classic users. Select a saved account in each workflow block; keep tokens out of node parameters. Resource IDs are text. Contact id and Company id are Intercom IDs; External id and External company id are your own identifiers. Create user or lead defaults to user and requires email or external ID for users; choose lead for an anonymous contact. Setting Email verified, including false, requires Email in the same request. Updates require at least one selected changed field and do not apply creation defaults. Timestamps use Unix seconds. False, zero and custom JSON attribute keys are preserved.

Contact lists and searches return one cursor page (at most 150 contacts); pass the returned cursor as Starting after. List defaults to ten and search to fifty. Search needs a nonempty Intercom filter object, for example {"field":"email","operator":"=","value":"rea@example.com"}. Search and company listing use POST but are read operations. List companies requests one numbered page with an explicit native default of twenty; Intercom limits page traversal to 10,000 companies. Find company by external company ID returns a list envelope. Attached contact companies use a companies array; most other lists use data. Read the returned envelope and page inputs yourself; the node never follows a returned URL automatically.

Set contact subscription preference requires an explicit Consent type, opt_in or opt_out, matching the subscription type. There is no consent default. Remove explicit contact subscription preference follows Intercom's preference-removal behavior; it is not a universal unsubscribe action. Review merges, deletions and consent changes before running. Deleting a tag accepts Intercom's empty HTTP 200 response. Provider permissions and workspace execution controls still apply, including WORKFLOW_LIVE_INTERCOM.

This port exposes core fields only. Conversations, attachments, Classic users, scroll traversal, triggers, OAuth renewal and automatic n8n parameter conversion are not included. Local contract, actual-control and saved-flow checks use simulated HTTP. Real Intercom US/EU/AU requests, permissions and hosted Review/organization setup remain unverified.

ActiveCampaign workflow nodes

Connect activeCampaignApi using ActiveCampaign API URL and ActiveCampaign API key from Settings → Developer. Copy the full HTTPS account origin, without /api/3, a query, fragment or embedded username/password. Do not construct it from an account name or assume a US hostname: the Developer tab is authoritative. Requests stay on the selected origin, use API v3 and send the key in Api-Token. Test connection reads /users/me; it does not certify writes, list permissions or feature eligibility. See authentication and base URL guidance.

The 38 operations cover current user, contacts, tags, contact lists, CRM accounts, account/contact associations, contact tags, list memberships, contact field definitions and stored field values. Select the saved account in each workflow block. Most IDs are positive decimal text strings; account/contact association write inputs, account owner and account custom-field IDs use safe positive integers. Remove contact tag by association ID needs the returned contactTag.id, not the tag definition ID. CRM account associations also have their own ID. A contact can belong to one CRM account; changing an association can move it and trigger account-change automations.

Create a new contact by email fails if that email already exists; Create or update contact by email uses /contact/sync instead. Both native forms require email; phone-only and WhatsApp lookup modes are not included. Updates change only selected fields. Contact Field values takes JSON objects such as [{"field":"61","value":0},{"field":"62","value":false}]; use existing field IDs. Numbers, booleans, null and string arrays retain their JSON types, with storage determined by the field type. Tags adds names without removing existing tags. Pass primitive arrays as JSON, for example ["Customer"]. Standalone Set contact custom field value and Update stored field value require a text value; checkbox/listbox text uses ||option 1||option 2||. Updating a stored value also requires its contact and field definition IDs. To send an empty text value, use a complete expression returning an empty string, such as {{input.value}}; blank editor fields are omitted.

Collections request one offset page, default twenty and at most one hundred. Pass Offset to read another page; contact lists also support Id greater and Order id for traversal. Exact tag lookup uses Exact name; Search is a substring search. Tags have separate contact/template types. Returned envelopes, side-loaded resources and meta.total are preserved; a returned link is never followed automatically. Get all memberships of one contact and Get all stored custom values of one contact return the provider's complete per-contact set and do not accept pagination, since these endpoints ignore limit/offset. See pagination.

Create contact list requires name, URL-safe string ID, sender URL and sender reminder. Channel defaults to email. Creating a list does not assign group visibility. Update contact list requires its name even when unchanged; omitted fields retain current values and the channel cannot change. Sender postal fields can be cleared using an expression returning empty text. Private is exposed on update, where false has its ordinary meaning; the provider's presence-based create behavior is not exposed. Tag names cannot contain commas, line breaks, HTML or four-byte characters; creating a tag explicitly sets its type, default contact.

Set explicit contact list status has no default: 0 unconfirmed, 1 active, 2 unsubscribed, 3 bounced. Changing status can resubscribe a contact and run the provider's automations, webhooks, autoresponders and notifications. Responder controls later scheduled autoresponders, not immediate subscribe mail. The separately named immediate-autoresponder and last-campaign options follow the provider's rules; omission preserves those rules. Review the subscription behavior before running. Successful membership writes must return the documented contactList envelope.

Execution uses WORKFLOW_LIVE_ACTIVECAMPAIGN and existing account/workspace controls. This port does not add deals, commerce/DeepData, campaigns, bulk imports, triggers, token issuance/renewal or automatic n8n conversion. Local protocol, actual-control and saved-flow checks use simulated HTTP. Real ActiveCampaign permissions, mail/automation effects and hosted Review/organization setup remain unverified.

NocoDB workflow nodes

Connect nocoDbApiToken with NocoDB host and NocoDB API token. Host defaults to https://app.nocodb.com; use the HTTPS origin of your cloud or self-hosted deployment, without an API path, query, fragment or embedded credentials. The deployment must support API v3. Requests use xc-token and stay on the selected origin, subject to existing outbound network rules. JWT/password login, token issuance and v1/v2 parameter conversion are not included. Test connection reads the v3 accessible workspace list; it requires the corresponding metadata permission and can fail for a valid data-only token. Success does not certify table writes, other scopes or plan availability. See API tokens and v3 REST API reference.

The 24 operations include workspace/base/table/view/field metadata reads, individual and bulk record create/update/delete, paged record reads, count, upsert, linked records, link/unlink, base64 cell attachments and button actions. Copy workspace/base/table/view/field IDs from NocoDB. Metadata and view API availability depends on the provider's plan and permissions; view APIs may require Enterprise on Cloud or a licensed self-hosted plan. These are reads of provider resources; they do not change Confluye organization permissions. See availability and ID guidance.

API v3 separates a record's id from its fields; it does not return the v2 flat row shape. Record id is a text primary key, including composite keys; keep long numeric-looking keys as text to preserve every digit. Scalar ID expressions from earlier nodes render as text, so numeric zero becomes "0", which NocoDB accepts. Bulk JSON IDs accept text or safe numeric values. Record fields use their exact titles or IDs and arbitrary JSON values: false, zero, null, arrays and nested keys are retained. Single-record creation accepts an empty field object when server defaults are sufficient; updates require at least one selected field.

Record and linked-record lists request one numbered page, default 25, with a native page-size limit of 1,000. The provider may silently apply a smaller deployment or Cloud cap, so inspect the returned pagination. Returned next, prev, nestedNext and nestedPrev URLs are retained; the node never follows them automatically. Read another page using Page, or Nested page and Nested limit for linked data embedded in record-list results. Field selection takes a JSON array such as ["Name","Count"]; sorting takes objects such as [{"field":"Count","direction":"desc"}]. Where uses the v3 filter syntax, for example (Count,eq,0). View id on list reads uses the view's visible fields/filters; explicit sort takes precedence over its sort. Field id on result and Links as ltar send their true-only flags when enabled and omit them when disabled. A linked-record result can contain records, a single record, or record:null depending on relationship kind.

List workspace bases requests one offset page and preserves pageInfo. Use the returned effective pageSize for the next offset, since the deployment can cap the requested limit. An offset past the available base list may return an error. Other native metadata list operations expose the provider's returned list without automatic traversal.

Bulk creates, updates, deletes and upserts accept one to ten records per native request; a deployment may set a smaller cap. Bulk create/update use JSON objects containing fields, with id also required for updates. Upsert records requires one to three Fields to merge on and a records array, for example [{"fields":{"Name":"REA","Count":0}}]. Multiple matching rows are rejected by the provider. Upsert display-value matching for links follows NocoDB's rules and may choose the first match; use known IDs when exact related records matter. The returned upsert status, computed-field warnings, ignored_fields and field_errors remain available for inspection. A successful HTTP response does not mean every requested computed field was written. Link/unlink add or remove specified relations; updating a link field directly replaces that record's link set. Deletion requires a positive deleted acknowledgment for every returned record, and link operations reject success:false. DELETE operations use JSON bodies; a proxy that removes those bodies needs a configuration change or another client.

Upload base64 attachment to a cell sends Content type, File (base64 text) and Filename to the selected record and field. It is not a local file picker or a remote-URL downloader. Run or preview a button action takes at most 25 text record IDs; Preview defaults to false, which runs the configured provider action. Both modes use write execution controls. Review the configured action and provider behavior before running. Execution uses WORKFLOW_LIVE_NOCODB and the saved account/workspace controls.

Metadata creation/deletion, membership administration, tokens, scripts, dashboards, provider workflows, comments, triggers and automatic n8n import compatibility are not included. Local contract, actual-control and saved-flow checks use simulated HTTP. Real NocoDB cloud/self-hosted permissions, plan eligibility, attachment storage, link changes and button effects, plus hosted Review/organization setup, remain unverified.

Kit (ConvertKit) workflow nodes

Connect kitApi with Kit API v4 key, created in the Developer settings of your own Kit account. Requests use X-Kit-Api-Key on https://api.kit.com/v4. Enter the v4 key in the named field; Advanced JSON uses {"apiKey":"…"}. ConvertKit v3 apiSecret/api_secret JSON and OAuth-token JSON are rejected by this credential type. Test connection reads current account identity and requires an API-eligible account; it does not certify writes or delivery. Kit documents this key mode for personal account automation; public apps and OAuth-only endpoints require a separate OAuth implementation. This port does not issue or renew tokens. See authentication.

The 40 native operations cover account/profile/statistics and brand colors, custom field definitions, forms, sequences, subscribers, filtering, locations and tags. The catalog name includes ConvertKit so it can be found by its former name. Replace account brand colors replaces the whole palette with up to ten codes; [] clears it. Native scalar IDs use positive decimal text; Kit returns numeric IDs on regular resources and text IDs in filter results. ID expressions render both forms as text. Reuse returned IDs and custom-field keys rather than constructing them.

Lists return one cursor page, with a native default of fifty and maximum of one thousand. Kit's own default is five hundred. Choose After with pagination.end_cursor when has_next_page is true, or Before with start_cursor; do not set both. Include total count can add provider query cost. Subscriber lists default to the provider's active-only view; choose status:all when appropriate. Forms have their own active/archived/trashed/all status and embed/hosted type filters. The node preserves the returned pagination and does not fetch further pages automatically. See pagination.

Create or update subscriber by email is an upsert. Its State applies to creation; it does not change an existing subscriber's state. Update subscriber requires Email address even when unchanged. Custom Fields use existing field keys and text values, for example {"counter":"0","flag":"false"}, with at most 140 keys. Unknown keys are not created by writing values: inspect errors and returned warnings. More than ten fields are processed asynchronously; HTTP acceptance and returned values do not prove the new fields are already stored. Read the subscriber later to verify them. There is no automatic polling or write retry. To send empty first-name text, use a complete expression returning ""; blank editor fields are omitted. Unsubscribe subscriber revokes future mail consent while retaining the subscriber record, history and tags; it is not a record deletion.

Form, sequence and tag additions require an existing subscriber. Create/upsert the subscriber first, then use the ID or email variant. Form additions can trigger Kit's incentive/opt-in behavior; sequence additions follow the sequence schedule and automations. The form ID variant sends Referrer, defaulting to empty text; set it when attribution matters. Tag removal removes that association and keeps the subscriber and other tags. This key-only port has no single-tag deletion operation; Kit's bulk deletion endpoint requires OAuth.

Create an empty sequence requires a name and uses the native default Active:false. Populate emails in Kit; sequence-email editing is outside this port. Updates require at least one changed field and apply no creation defaults. Send hour accepts zero for midnight in the sequence's time zone. Time zone uses an IANA name, and Send days uses lowercase weekday values. Explicit false Active, Hold and Repeat, plus empty day/exclusion arrays, are retained. Repeat can restart a subscriber's sequence position when they re-enter; hold keeps them eligible for future published emails. Exclusion source types are tag, sequence, form and segment, with numeric source IDs. Renaming a custom field changes its key and can break Liquid personalization references; deleting a definition removes its subscriber values.

Filter subscribers uses POST as a read operation. Supply All conditions combined with AND, for example [{"type":"subscriber_state","states":["active"]},{"type":"opens","count_greater_than":0}]. Engagement, signup, state, tag, attribution, custom-field and location conditions have typed fields. Unknown condition keys are rejected. State filters require states; location filters require coordinates and radius in miles; custom-field filters need an ID/value except has_value; tag and attribution filters need applicable any conditions. The counting mode selects raw events or unique emails. Optional includes retain tags, statistics, fields, location and attribution as returned. Kit's historical data limits still apply; see email data retention. Location pinning takes a JSON object containing city, state_province, uppercase two-letter country_code, numeric latitude/longitude and IANA timezone; zeros and empty state text are valid native values.

API-key rate limits and plan permissions apply, along with WORKFLOW_LIVE_KIT and existing account/workspace controls. Bulk OAuth jobs, purchases, broadcasts, sequence emails, webhook triggers and automatic n8n parameter or credential conversion are not included. Local contracts, saved-flow and actual-control checks use simulated HTTP. Real Kit account permissions, async completion, incentive/email delivery, sequence effects and hosted Review/organization setup remain unverified.

Bitly workflow nodes

Bitly blocks provide 33 API v4 operations for links, custom aliases, groups, branded-domain reads and analytics. Create a saved bitlyApi account using Bitly access token from Bitly Settings > API, then use Test connection and select that account on the block. Advanced JSON accepts {"accessToken":"YOUR_TOKEN"}; enter the token alone in the named field. The connection test reads /v4/user; it verifies identity, not write or analytics permissions. This form does not issue or renew OAuth tokens.

Use the returned text Bitlink ID, such as bit.ly/abc123, including the domain and slash. A full https:// link is not an ID. Custom aliases use the same domain/keyword format. IDs remain text, including numeric-looking hashes, and can be mapped from an earlier block. All requests use https://api-ssl.bitly.com/v4; a node cannot replace the API host.

Shorten link requires an HTTP or HTTPS destination. Create link additionally accepts title and tags; optional group/domain fields use the selected account's permissions. Update link requires at least one changed destination, title, archive flag or tag list. Explicit Archived:false unarchives the link; an empty tags array clears its tags. An empty title can be supplied with an expression returning ""; blank editor fields are omitted. Updates do not apply creation defaults. Delete link is permanent and subject to Bitly restrictions. Creating/updating custom aliases can redirect existing traffic; review the target before running. Expand link uses POST as a read operation.

List group links returns one page with native Size:50 and an optional Search after token. Reuse the API's returned pagination.search_after token; pagination URLs are not followed automatically. Optional created-before/after filters use integer Unix seconds, including zero. List top group links sorts by clicks and permits at most 100 entries. Metrics use Unit:day and Units:-1 by default; -1 requests all available history, while a positive number selects that many units. Optional Unit reference is a timestamp; enter its literal + and : characters and the node encodes them. Available history, metrics, redirects, custom domains and quotas depend on the Bitly plan.

WORKFLOW_LIVE_BITLY and existing account/workspace controls govern execution. Group/user administration, bulk changes, QR-code creation, domain registration, billing, deeplinks, dynamic-routing authoring and webhook triggers are outside this port. Automatic n8n parameter conversion is not included. Local contract, saved-flow and control checks use simulated HTTP; real Bitly permissions, redirects, deletion, analytics availability and hosted Review/organization persistence have not been verified without a provider account and Review deployment.

Strapi workflow nodes

Strapi blocks provide 14 Strapi 5 Content REST operations: collection documents, single types, explicit publication and media-file reads/deletion/metadata updates. Create a saved strapiTokenApi account with its HTTPS origin, scoped Content API token, API prefix (default /api) and Collection for connection test. Use the plural API ID of a collection the token can read, such as articles. The origin has no path; put a custom deployment/REST path in the saved API prefix. Advanced JSON uses {"url":"https://cms.example.com","apiToken":"YOUR_TOKEN","apiPrefix":"/api","testCollection":"articles","apiVersion":"v5"}. This account does not accept password login, Admin tokens or Strapi v3/v4 credentials.

Test connection reads at most one document from that test collection with counting disabled. It verifies that read, not every content type, media action or write. The token's role, scopes and expiry still apply. Select the saved account on each block. A node cannot override its host or API prefix.

For a collection, Content type is its plural API ID, such as articles; for a single type, use its singular API ID, such as homepage. Document id is the returned text data.documentId, not the numeric database id. Numeric-looking IDs stay text. Responses keep the Strapi 5 flattened data envelope; the port does not convert old data.attributes responses.

Supply exact Data fields as a JSON object. False, zero, null, empty text and empty arrays are retained, including nested components or relation connect/disconnect/set objects supported by your schema. Strapi validates application-specific fields and permissions. Native content writes default to Status:draft, explicitly different from the REST API's default of publishing immediately. Drafts require Draft & Publish on that content type; a type without that feature cannot be made draftable by this flag. Choose published when appropriate. Publish document and Publish single type publish the existing draft using an empty data object and explicit published status. Deleting content is permanent; an optional locale scopes the request and Strapi determines which locale/version records are removed.

Reads default to Status:published and support locale, nested filters/populate, field selection and sorting. For example, Filters {"$and":[{"count":{"$gte":0}},{"active":{"$eq":false}}]} keeps the numeric and boolean values while using Strapi's indexed bracket query format. Fields and Sort use JSON arrays such as ["title","count"] and ["title:asc"]. Populate accepts * or a JSON object. Population needs the relevant read permission. Queries support at most twelve nested levels in the native adapter.

Collection and media lists return one page, using Page:1, Page size:25, With count:true. Set With count to false to avoid the server count query. Strapi's configured default/max limits may cap results; increase Page to continue. Media lists use the current paginated /upload/files/page endpoint, which older Strapi 5 deployments may not expose. Media IDs accept a text numeric ID or the returned file documentId; files do not have draft/published versions. Update media file info accepts name, alternativeText, caption and nullable focalPoint:{"x":0,"y":0} with coordinates from 0 to 100. Empty text and zero coordinates are retained. Strapi defaults an omitted focal point to null, so include the existing point when preserving a crop. Metadata updates do not replace the binary file.

WORKFLOW_LIVE_STRAPI and existing account/workspace policy govern execution. Binary upload/replacement, schema/admin management, password authentication, automatic n8n/old-version parameter conversion and unbounded media listing are outside this port. Local contract, saved-flow and actual-control checks use simulated HTTP. Real Strapi schema validation, token scopes, publication, localized deletion, media persistence and hosted Review/organization setup remain unverified without a configured Strapi server and Review deployment.

Clockify workflow nodes

Clockify blocks provide 38 API v1 operations for workspaces, users, clients, projects, tasks, tags and time entries. Create a saved clockifyApi account with Clockify API key from Profile Settings and its API region: global, eu (Germany), us, uk or au. Subdomain workspaces need a key generated specifically for that workspace. Advanced JSON accepts {"apiKey":"YOUR_KEY","region":"eu"}; the named form defaults its region to global. Test connection reads the selected key owner's identity. It does not certify workspace roles, writes or paid features. API requests stay on the selected regional Clockify host; a block cannot override the API URL.

Workspace, user, client, project, task, tag and entry IDs are text. Select a workspace explicitly; account region does not choose it. Lists return one page using Page:1, Page size:50 and the provider's remaining filters. Use JSON arrays for multi-value filters, such as Clients ["client1","client2"]; they are sent as separate repeated query parameters. Empty query arrays omit the filter; write arrays are sent explicitly. Workspace-role filtering takes one role, while Search workspace users accepts a role array. User search uses POST as a read. Native results retain the API's JSON shape; automatic pagination and response-header pagination markers are not added. Advance Page to request another page.

Create project and Create project from template require explicit Visibility:public/private. Private projects require an eligible paid plan; the node does not choose visibility for you. Project/client/tag updates require at least one changed field, and task updates require the task name. Native updates forward the supplied fields without adding creation defaults. Clockify PUT semantics still apply: read the existing record and include values you need to retain, then inspect the returned record. Explicit false archive, billing, template and estimate flags are retained, as are empty assignee/tag arrays, zero rate/estimate values and midnight reset hour. Project rate amounts use Clockify's raw integer API values; the node does not convert display currency amounts. Nested estimate, membership, task and custom-field JSON follows Clockify's current schema and permissions.

Time-entry creation and updates require an explicit UTC Start timestamp, for example 2026-10-17T10:00:00Z. Supply End for a completed interval; omitting End when creating an entry starts a running timer. Stop user timer requires the target user and an explicit End and can affect that user's active timer when the key has permission. Create user time entry also targets the selected user. For an empty description or to clear text associations where the API permits it, use an expression returning ""; blank editor fields are omitted. Use a whole-array expression to send [] explicitly. Time-entry deletion consumes an empty HTTP204 response; other entity deletions return the provider's deleted-resource object.

Add workspace user changes membership and defaults Send email:false; enable it explicitly to request invitation mail. The operation returns a workspace object, not a new user ID. Owners/admins and other permitted roles still control access. Project visibility, estimates, rates, user changes and other features remain subject to Clockify plan limits. These adapters do not bypass approvals or restrictions on time entries.

WORKFLOW_LIVE_CLOCKIFY and existing account/workspace policy govern execution. Reports, expenses, invoices, time off, approval workflows, bulk entry changes, role administration, add-on tokens and webhook triggers are outside this batch. Automatic n8n parameter conversion is not included. Local contract, saved-flow and actual-control checks use simulated HTTP. Real key/subdomain permissions, PUT omission/clearing behavior, timers, invitations, plan restrictions and hosted Review/organization persistence remain unverified without a Clockify account and Review deployment.

Webflow workflow nodes

Webflow provides 33 Data API v2 operations for sites, custom domains, publishing, collections, staged and live items, page metadata, asset metadata/folders, forms/submissions and webhook reads. Save a webflowApi account using Webflow access token, or valid Advanced JSON {"accessToken":"YOUR_TOKEN"}. Use a v2 site token or an existing OAuth access token with the required resource scopes. Test connection reads /v2/sites and requires sites:read; it does not certify CMS writes, publishing or other permissions. Requests stay on https://api.webflow.com; the node cannot override the host. Token issuance and automatic renewal are not provided.

Use 24-character resource IDs as text. Create staged item requires fieldData.name and fieldData.slug, defaults Is draft:true, and forwards other field values as JSON. Staged changes and live changes use separate operations. Update live item, Unpublish item, and Publish selected staged items affect the live site; inspect publishedItemIds and errors because a successful HTTP response can include publishing errors. Publish site or page requires at least one custom domain ID or explicit publishing to the Webflow subdomain; publishing can expose staged changes. The response acknowledges the request and does not prove the public site finished publishing.

Paged lists return one page, default limit 100 and offset 0; use the returned pagination to request another page. Site/collection/folder/webhook listings use their documented unpaged endpoints and return their raw response metadata. CMS locale IDs and page/asset locale IDs are separate inputs. Set Locale ID on an asset update to target translated alt text; the asset display name is not localizable. Updates require a changed field and retain explicit false, zero and empty JSON field values. Asset metadata updates accept an explicitly mapped empty string to clear alt text; an empty optional text box is omitted by the editor. Existing provider rules still govern required values, field types and permissions.

WORKFLOW_LIVE_WEBFLOW and workspace/account policies govern execution. This batch does not add binary asset upload, CMS field administration, live-item creation, bulk item changes, rich page editing, e-commerce, webhook registration/trigger processing or n8n parameter conversion. Local protocol, saved-flow and actual-control checks use simulated HTTP. Real token scopes, CMS validation, live publishing/unpublishing, asset changes, hosted Review Help and organization persistence remain unverified without suitable service credentials and a Review deployment.

Help Scout workflow nodes

Help Scout provides 26 Mailbox API v2 operations for conversations, customers, conversation threads, mailboxes, users and tags. Save a helpScoutOAuth2Api account with Help Scout access token, or valid Advanced JSON {"accessToken":"YOUR_TOKEN"}. Supply an existing OAuth access token; these nodes do not issue or refresh tokens. Replace expired tokens. Test connection reads /v2/users/me and only checks the selected token's identity. Requests stay on https://api.helpscout.net.

Use positive decimal resource IDs as text within the API's safe integer range. Lists return one server page with HAL links and pagination; pass the next page number explicitly. Links are returned as data and are not fetched automatically. Conversation creation and reply operations can send customer email; review recipients, mailbox permissions and write effects before running. Creation returns the Resource-ID and Location response headers as data rather than inventing a JSON entity body. Customer overwrite and tag/custom-field replacement can replace existing values; read the current record first. Thread import dates require the corresponding imported flag. Runtime validation enforces customer identity and compatible patch path/value combinations.

WORKFLOW_LIVE_HELPSCOUT and workspace/account policies govern execution. OAuth consent/renewal, triggers and unbounded pagination are outside this batch. Local contract, saved-workflow and actual-control checks simulate HTTP; live mailbox permissions, email delivery, overwrite behavior, hosted Review Help and organization persistence remain unverified without a Help Scout account and Review deployment.

Mautic workflow nodes

Mautic provides 28 REST API operations covering contacts and companies, segment/campaign/email reads, contact relationships, membership changes, points and email sends. Save a mauticApi account using Mautic HTTPS origin, Mautic username and Mautic password. Advanced JSON accepts {"baseUrl":"https://mautic.example.com","username":"YOUR_USER","password":"YOUR_PASSWORD"}. Enable both the API and HTTP Basic Auth in Mautic and grant this user the needed permissions. This adapter requires an HTTPS origin without an installation path. It does not accept OAuth credentials or disable TLS verification.

Test connection requests one contact page at /api/contacts?limit=1&start=0. This checks contact-read permission, not sending or mutation permissions. Blocks cannot change the selected origin. Use positive decimal IDs as text within the safe integer range and exact contact/company field aliases. Lists return one page using start/limit; totals can be numbers or decimal text. Zero, false and explicit empty values survive JSON field mapping. Contact/company updates require Overwrite with blank:true when deliberately clearing existing fields. HTTP 200 error envelopes are rejected. Membership changes, points adjustments and email sends are writes and can trigger Mautic automation or outbound mail; inspect provider results and do not assume delivery from acceptance alone. Segment-email sends target the segments already associated with that email in Mautic.

WORKFLOW_LIVE_MAUTIC and workspace/account policies govern execution. OAuth issuance, subdirectory installations, triggers and unbounded pagination are outside this batch. Local protocol, saved-flow and actual-control checks simulate HTTP. Real server/version compatibility, role restrictions, sends, automation side effects, hosted Review Help and organization persistence remain unverified without a Mautic account and Review deployment.

Coda workflow nodes

Coda provides 27 operations for account identity, docs, page metadata, tables/views, columns, rows, controls, formulas, buttons and mutation status. Save a codaApi account with Coda API token, or valid Advanced JSON {"accessToken":"YOUR_TOKEN"}. The current official API specification uses https://docs.superhuman.com/apis/v1; these nodes bind the selected token to that host and do not follow legacy API redirects. Test connection reads /whoami and does not certify document access or writes.

Keep resource IDs as text. Table/row/page identifiers that accept names are encoded as a single path segment; URLs and encoded paths are rejected. Lists return one bounded page; pass the returned nextPageToken explicitly. Use either a page token or a sync token where supported, never both. Returned links are data, not automatic follow-up requests. Row writes and page mutations can be asynchronous: inspect requestId using Get mutation status before treating them as complete. Button pushes can execute document automation. Page create/update operations cover metadata, not rich page content. Explicit false, zero and empty row values are retained.

WORKFLOW_LIVE_CODA and workspace/account policies govern execution. OAuth issuance/renewal, page content editing, triggers and unbounded pagination are outside this batch. Local contract, saved-flow and actual-control checks simulate HTTP. Real token/document permissions, asynchronous mutations, button effects, hosted Review Help and organization persistence remain unverified without a Coda account and Review deployment.

Monday.com workflow nodes

Monday.com provides 47 native operations for account identity, boards, columns, groups, items, updates, users and workspaces. Save a mondayComApi account using Monday.com API token, or valid Advanced JSON {"accessToken":"YOUR_TOKEN"}. Use a personal API token or an existing app access token with the permissions needed by each operation. Requests use the fixed https://api.monday.com/v2 GraphQL endpoint and explicitly pin API-Version:2026-07, a supported stable version in the official versioning documentation. Blocks cannot supply an arbitrary GraphQL query or override the API host/version.

Test connection runs a read-only me query and requires the app token's me:read scope when applicable. It does not certify access to particular boards, writes or plan features. Personal tokens inherit the account's permissions. App tokens need the documented resource scopes, including boards, updates, users and workspaces as applicable. Token issuance and automatic renewal are not provided.

Keep IDs as text. Paged operations return one bounded page; pass the returned cursor or page input explicitly. Column values use an object keyed by column IDs and the format required by each Monday column type; zero, false, null and empty values are retained. Change multiple column values requires at least one supplied column. API validation and GraphQL errors reject the operation rather than accepting partial data as success. Duplicate-board results can arrive before all copied content is ready; inspect the resulting board before depending on completion. Item/group/board deletions and updates affect real board content, and updates can notify subscribers.

WORKFLOW_LIVE_MONDAY and workspace/account policies govern execution. Arbitrary GraphQL, binary upload, OAuth renewal, trigger execution and automatic all-page collection are outside this adapter. Local protocol, saved-flow and actual-control verification uses simulated provider HTTP. Real token scopes, column formats, notifications, asynchronous board copies, hosted Review Help and organization persistence remain unverified without a suitable Monday account and Review deployment.

Calendly workflow nodes

Calendly provides 40 native REST operations for users, organizations and memberships, event types, availability, scheduled events/invitees, no-shows, scheduling links, routing forms, groups and teams. Save a calendlyApi account using Calendly access token, or valid Advanced JSON {"accessToken":"YOUR_TOKEN"}. Use a personal token or an existing OAuth access token. Requests stay on https://api.calendly.com; API URLs cannot be overridden. Test connection reads /users/me and requires users:read. Each operation still needs its documented scopes, including event, availability, organization or scheduling-link scopes as applicable.

Use text UUIDs for path resources and canonical https://api.calendly.com/... URIs where a node asks for a user, organization, event type or group URI. The operation validates the expected resource family; a general URL is not interchangeable. Event-type listings require a user or organization; scheduled-event listings require a user, organization or group. Time windows use ordered UTC timestamps. Available event-type slots require a future window of at most 31 days; user busy times require a future window of at most seven days. Lists return one page and opaque continuation tokens; returned page links are data and are not followed automatically.

Creating an event invitee books a meeting. Canceling an event, recording/removing no-shows, changing event types, issuing organization invitations and creating scheduling links have external effects and may notify other people. Check selected resources and permissions before running. Explicit false activation flags and empty descriptions are preserved; updates require a changed field. To clear a description through the native editor, use an expression that returns an empty string (for example {{input.description}} with an empty input value); a blank optional text box is omitted on save. Returned resources and acceptance do not prove downstream delivery or automation completion.

WORKFLOW_LIVE_CALENDLY and workspace/account policies govern execution. OAuth renewal, webhook-trigger execution, binary upload and automatic all-page collection are outside this adapter. The n8n reference exposes a Calendly trigger; native REST actions do not implement that trigger. Local contract, saved-flow and actual-control checks simulate HTTP. Real booking/cancellation, role/plan restrictions, notifications, hosted Review Help and organization persistence remain unverified without a Calendly account and Review deployment.

Typeform workflow nodes

Typeform provides 24 native REST operations for forms, form messages, responses, webhook definitions, workspaces and themes. Save a typeformApi account using Typeform access token, or valid Advanced JSON {"accessToken":"YOUR_TOKEN"}. Use a personal token or existing OAuth access token. Requests stay on https://api.typeform.com; EU API hosts are not supported by this adapter. Test connection reads one form page and requires forms:read; other actions require their own form, response, webhook, workspace or theme scopes. Token issuance and automatic renewal are not provided.

Keep IDs and response tokens as text. Response entries may contain response_id, token, or both; the adapter preserves whichever identifiers the provider supplies. Safe numeric theme IDs are normalized to text. Lists return one bounded page. Response pagination uses before or after, never both; do not follow returned links automatically. Partial form changes expose the supported title/public/workspace/theme inputs. Replace form sends a complete definition and requires the full fields array, including existing field IDs: omitted fields and their response data are deleted. Read the current form and retain fields that must survive. Custom-message updates require a nonempty object. Theme updates replace supplied color/field/screen objects, which must include all required subfields; public themes cannot be updated.

Response deletion is queued: API acceptance does not mean that deletion finished. Webhook operations manage definitions; they do not install a Confluye Typeform trigger or validate incoming events. Webhook signing secrets are not accepted in node parameters or exposed in the typed webhook output. Explicit false flags and supported empty values are retained. Review public-form changes, replacement and deletion effects before running.

WORKFLOW_LIVE_TYPEFORM and workspace/account policies govern execution. Binary upload, webhook signing-secret management, trigger execution, OAuth renewal and automatic all-page collection are outside this adapter. The n8n reference has a Typeform trigger only; its execution remains unimplemented. Local protocol, saved-flow and actual-control checks simulate HTTP. Real permissions, public form changes, response deletion completion, webhook delivery, hosted Review Help and organization persistence remain unverified without a Typeform account and Review deployment.

Shopify workflow nodes

Shopify provides 37 native Admin GraphQL API operations for shop identity, products, variants, inventory, locations, customers, orders, collections, fulfillment metadata and asynchronous job status. Save a shopifyAccessTokenApi account using Shopify shop domain and Shopify access token, or valid Advanced JSON {"shopDomain":"your-shop.myshopify.com","accessToken":"YOUR_TOKEN"}. Supply the canonical single shop subdomain, without HTTPS, port or path. The existing Admin access token is bound to that shop; nodes cannot override its origin. Requests pin API version 2026-10. Legacy REST Basic Auth, OAuth issuance and automatic renewal are not implemented.

Test connection reads shop identity and verifies its returned myshopifyDomain; it does not certify resource scopes. Products, variants and collections require read_products or write_products; inventory and locations need their own scopes. Customer reads/writes can require protected customer data approval. Order access normally covers the last 60 days; older orders need the applicable approval and read_all_orders. Fulfillment-order metadata access depends on the ownership scopes granted to the app. Consult each operation's current Shopify requirements.

Keep full typed gid://shopify/... IDs and decimal price strings as text. Lists return one bounded cursor page; pass the returned cursor explicitly. Fixed authored queries use variables, and GraphQL errors or mutation userErrors fail the operation even with HTTP 200. Updates require a changed field and preserve supported empty strings, false and zero. Tags supplied in an update replace the existing list; use Add tags or Remove tags for incremental changes. Newly created collections remain unpublished, and changing product status does not publish it to a sales channel. Collection writes also depend on an eligible plan. Product, customer and collection deletion is permanent. Inventory adjustment changes available quantity by an integer delta and requires an idempotency key for the exact intended adjustment; reuse it only for the identical intent. The returned adjustment reason is descriptive text and its changes list can include multiple inventory states, such as available and on_hand, for each requested adjustment; do not interpret that list length as the number of adjusted items. The adapter does not automatically retry uncertain writes. Collection product removal returns an asynchronous job; inspect Get asynchronous job status and the collection before assuming removal finished.

WORKFLOW_LIVE_SHOPIFY and workspace/account policies govern execution. Payment/refund operations, actual fulfillment changes, binary/media upload, arbitrary GraphQL, OAuth renewal, automatic all-page collection and webhook-trigger execution are outside this adapter. Native service actions do not implement the n8n Shopify trigger. Local protocol, saved-flow and actual-control tests simulate HTTP. Real app scopes, protected data access, inventory/product/customer effects, asynchronous completion, hosted Review Help and organization persistence remain unverified without a suitable Shopify account and Review deployment.

Salesforce workflow nodes

Salesforce provides 11 native REST API operations parameterized across Account, Contact, Lead, Opportunity, Case and Task. They cover object metadata, record reads and structured lists/counts, create/update/delete, external-ID upsert/lookup and supported parent relationships. Save a salesforceOAuth2Api account using Salesforce instance URL and Salesforce access token, or valid Advanced JSON {"instanceUrl":"https://your-domain.my.salesforce.com","accessToken":"YOUR_TOKEN"}. Use an HTTPS My Domain origin, including supported sandbox subdomains, without port, path, query or embedded credentials. Tokens remain bound to that selected instance. Requests pin supported REST version 66.0; this is not a latest-version claim.

Use an existing OAuth access token with api scope and the user's API Enabled permission. Object CRUD, field permissions and record sharing still apply. Test connection reads /services/data/v66.0/sobjects and checks accessible object metadata; it does not prove access to a particular record or write. JWT signing, token issuance and automatic renewal are not provided. The n8n JWT credential mode is not accepted as an alias for an existing OAuth access token.

Keep 15- or 18-character Salesforce IDs as text. Field maps accept scalar values or null, retain false/zero, and require 1–100 changed fields without Id or attributes. Null clears a field; use text for full-precision decimals. Fractional numbers and unsafe integers returned in raw Salesforce JSON are preserved as their original text to avoid precision loss; safe integer counts, zero and booleans retain their native types. Structured record queries accept field names and bounded filters, without arbitrary SOQL. Lists return at most 200 records per page. To continue, reuse the returned version-specific relative nextRecordsUrl as Cursor, keep the same object, and omit fields, filters, ordering and custom limit. Returned URLs cannot select another host or API version. An explicit query limit caps the whole query rather than requesting unlimited later pages. Related-record reads cover the documented parent relationships only. Prefer a unique external ID for upsert; a nonunique field can require View All Data. Set Update only:true to prevent insertion when an external ID is absent.

WORKFLOW_LIVE_SALESFORCE and workspace/account policies govern execution. Writes may run Salesforce automation or notify people; deletion changes real CRM records. Acceptance does not certify downstream automation. Custom objects, bulk/async jobs, attachments, arbitrary SOQL, triggers and OAuth/JWT lifecycle management are outside this adapter. Local protocol, saved-flow and actual-control checks simulate HTTP. Real instance access, sharing/field permissions, upsert/delete effects, automation, hosted Review Help and organization persistence remain unverified without a Salesforce account and Review deployment.

Zoho CRM workflow nodes

Zoho CRM provides 22 native CRM API v8 operations for records, search, bounded batch writes, module/field/related-list metadata, users, organization and notes. Save a zohoOAuth2Api account using Zoho CRM region and Zoho CRM access token, or valid Advanced JSON {"region":"us","accessToken":"YOUR_TOKEN"}. Choose the token's matching region: us, eu, in, au, cn, jp, ca or sa. The selected account binds the corresponding official www.zohoapis origin and fixed /crm/v8 path; nodes cannot replace that host. Other regions and sandbox/developer origins are unsupported. Use an existing OAuth access token; issuance and renewal are not provided.

Test connection reads the current user and requires ZohoCRM.users.READ. Other operations require the matching module, settings, organization or user scopes; search also requires ZohoSearch.securesearch.READ. CRM role, profile, layout and field permissions remain authoritative. A successful check does not prove write access.

Keep decimal resource IDs as text, including nested lookup IDs. Use exact module and field API names from metadata. Read field selectors also accept the documented $approval_state and $event_cancelled system fields; these exceptions do not change module or related-list path names. Lists return one explicit page, normally at most 200 records; use either a page number or the returned account/query-bound page token. Search requires exactly one mode: email, phone, word or criteria. Provider search matching can be broader than exact equality, so inspect returned fields before a write. Conditional reads preserve the supplied modification timestamp; an unchanged HTTP 304 is reported as an explicit failure without retry under the shared gateway policy, while supported HTTP 204 empty reads return their no-content metadata. Batch writes accept at most 100 nonempty field maps; updates need each record's text id and at least one changed field. Upsert uses unique fields and excludes activity modules. Explicit empty strings, false, zero and null are retained.

Writes may trigger CRM automation. An empty Trigger array suppresses workflow rules, while omitted Trigger uses provider defaults; other configured automation such as cadences can still execute. Deletion workflow execution has a separate boolean. Mixed per-record failures are reported as failures rather than certifying the entire batch; successful items may already have changed and are not rolled back. Read the records before retrying an uncertain write. Notes can be shared with a linked portal user when explicitly requested. A blank optional text box is omitted by the editor; to clear a note's supported text field, map an expression returning an empty string.

WORKFLOW_LIVE_ZOHOCRM and workspace/account policies govern execution. OAuth lifecycle management, triggers, arbitrary COQL, binary files, asynchronous bulk jobs and automatic all-page collection are outside this adapter. Local protocol, saved-flow and actual-control checks simulate HTTP. Real regional permissions, mandatory fields, batch persistence, automation/portal sharing, hosted Review Help and organization persistence remain unverified without a Zoho CRM account and Review deployment.

Dropbox workflow nodes

Dropbox adds 25 native API v2 JSON RPC operations for account/storage information, file/folder metadata, bounded lists/search/revisions, folder creation, copy/move/delete/restore, shared links, folder sharing and share-job status. Save a dropboxOAuth2Api account using Dropbox access token and optional Dropbox namespace ID / Dropbox selected member ID, or Advanced JSON {"accessToken":"YOUR_TOKEN","namespaceId":"12345","selectUser":"dbmid:MEMBER_ID"}. Omit optional fields for a normal personal token. The token uses the fixed api.dropboxapi.com/2 API. Node inputs cannot override its origin, token, saved routing headers or selected member. Existing access tokens are used without issuance or renewal.

Test connection posts JSON null to /2/users/get_current_account with account_info.read and the stored selected-member header, if supplied. It reads identity only and does not verify namespace, file or sharing access. A team-token member selection requires team_data.member and the corresponding team app permissions. The decimal namespace ID supplies default path routing, rather than an access-control boundary. Supported explicit file IDs and ns:ID/path read paths can address other resources permitted to the same Dropbox principal. Dropbox permissions and token scopes determine access. Keep namespace IDs as text and use the path/ID formats described by each operation. A continuation cursor belongs to the same account and query. Each call returns one bounded page; search has its own 1,000-result request and 10,000-result overall limits, and results can lag or repeat. No server-provided link is automatically followed.

Read operations use POST where required by Dropbox RPC; POST alone does not indicate a write. Use files.metadata.read, files.content.read/files.content.write and sharing.read/sharing.write as required by the selected operation. Updates preserve supported false, zero and empty values. Moving, deleting, restoring, creating/revoking shared links and sharing folders are writes. Share folder can create a missing folder and change inherited access; full Dropbox access can be required. An async job ID or in-progress result is pending, so call Check share-folder job status explicitly. Writes are not retried automatically; inspect the provider state before repeating uncertain work.

WORKFLOW_LIVE_DROPBOX and workspace/account policies govern execution. Binary upload/download, OAuth lifecycle, automatic pagination, team administration and trigger execution are outside this port. Local protocol, saved-flow and actual-control tests simulate HTTP. Real namespace/team permissions, file changes, shared access and asynchronous completion, hosted Review Help and organization persistence remain unverified without a Dropbox account and Review deployment.

Box workflow nodes

Box adds 29 native REST API operations for file/folder information, creation/copy/update/trash, search, comments, tasks, collaborations, shared links and metadata instances/ordered JSON Patch. Save a boxOAuth2Api account using Box access token, or Advanced JSON {"accessToken":"YOUR_TOKEN"}. Requests use the fixed https://api.box.com/2.0 endpoint with the selected existing token. Node inputs cannot override its host or authorization. Token issuance and renewal are not implemented.

Test connection reads /2.0/users/me and verifies a user identity. File and folder operations need the matching application read/write permissions and the selected user's access; the check does not certify those resources or writes. Keep decimal IDs as text, including the root-folder ID 0. Parameterized item operations require choosing file or folder. Lists use one page with the local limit of 100 entries; use either marker or offset where available. Task lists do not offer pagination. Updates require at least one changed field and preserve supported empty strings, false and zero.

Comments, tasks and collaboration invitations can notify people. Collaboration changes can grant, revoke or transfer access; a new collaboration needs exactly one collaborator ID or user email, and groups use IDs. Shared-link operations can expose content under the chosen access/permission settings; editing requires a file with downloads explicitly enabled. Recursive folder deletion moves descendants to trash. Edit a metadata patch as a complete JSON array in its wrapping text editor. Metadata patches execute in order: add/replace/test require a value; move/copy require a source pointer; remove uses its target pointer. System metadata fields cannot be patched. HTTP 204 means a supported empty completion; unexpected statuses, malformed envelopes and incomplete results fail without automatic write retries.

WORKFLOW_LIVE_BOX and workspace/account policies govern execution. Binary transfers, OAuth lifecycle, automatic all-page collection and the n8n Box trigger are outside this port. Local HTTP is simulated. Real app/user permissions, notifications, shared-link exposure, ownership transfer, trash and metadata persistence, hosted Review Help and organization persistence remain unverified without a Box account and Review deployment.

Zoom workflow nodes

Zoom adds 14 native API v2 operations for user identity, meetings, invitations, registrants and recording metadata. Save a zoomOAuth2Api account using Zoom access token, or Advanced JSON {"accessToken":"YOUR_TOKEN"}. Use an existing OAuth token against the fixed https://api.zoom.us/v2 API. JWT signing, OAuth issuance and automatic renewal are not implemented. These are native service capabilities; registrant/recording actions extend beyond the Meeting resource exposed by the inspected n8n package.

Test connection reads /v2/users/me with granular user:read:user or the applicable classic user-read scope. Meeting, registration and recording operations need their own current scopes and host/account permissions. User-level apps normally use me; selecting other users requires suitable authority. Licensed meeting registration and Pro-or-higher cloud recording features depend on the host. A successful identity check does not prove those features or write scopes.

Use meeting numbers as exact decimal text and raw recording UUIDs where requested; special UUIDs are encoded by the adapter. Reads return one page of at most 300 records with explicit continuation tokens. Recording list windows have a conservative local maximum of 28 days. Recording output covers the current documented completed-file metadata and omits media download/access tokens and passcodes; unknown statuses fail explicitly. Host start URLs are omitted. Participant join URLs and invitation instructions can still be sensitive, so choose where workflow output is shared.

Creating/updating/deleting meetings and changing registrations are writes and can notify attendees. Creating meetings supports instant and scheduled types; recurring scheduling is not implemented. Deleting without an occurrence ID can delete an entire recurring series. Local start times require a valid timezone; updates need at least one changed meeting field and preserve false, zero and supported empty text. Writes are never retried automatically. WORKFLOW_LIVE_ZOOM and workspace/account policies govern execution.

Binary recording downloads, account/billing administration and webhook triggers are outside this port. Local protocol, saved-flow and actual-control tests simulate HTTP. Real licenses, scopes, scheduling/registration effects, notifications and recording availability, hosted Review Help and organization persistence remain unverified without a Zoom account and Review deployment.

Discourse workflow nodes

Discourse provides 21 native operations for categories, posts, topics, groups, user profiles, private-message lists and search. Save a discourseApi account with Discourse site URL, Discourse API key and Discourse API username, or Advanced JSON {"baseUrl":"https://forum.example.com","apiKey":"YOUR_KEY","apiUsername":"your_user"}. This port supports public, root-mounted HTTPS sites on port 443. Subfolder installations, private hosts and Unicode acting usernames are unsupported. Supported usernames have 1–60 ASCII letters, digits, dots, underscores or hyphens and start with a letter, digit or underscore. Nodes cannot change the saved host, key or acting user.

Test connection reads /session/current.json with the saved Api-Key and Api-Username, checks the returned user ID and matches the saved username. Restricted keys need permission to read this session endpoint; public category/group lists are not an authentication check. The check does not establish write or moderation permissions. Deployed server versions and site settings vary; the inspected current API reference does not pin your site's version.

Use exact positive text IDs. Each read returns one bounded page or the initial topic post chunk; it does not automatically retrieve every post. Create topic, reply or private message requires a mode: topic needs a title, reply needs topic_id, and private message needs a title and existing recipient usernames. Group, category and moderation operations require the corresponding site authority. Posts, messages, edits and group membership changes may notify users or send email. A queued post produces an explicit publication-unconfirmed error, with no automatic retry. Inspect Discourse before repeating any uncertain write. Enabling a topic pin requires a future date in this supported subset. To clear a user display name, turn on Clear Name and omit Name; otherwise enter a nonempty Name. Blank optional profile fields are omitted when saved; use an expression that renders an empty string when an operation explicitly supports clearing them.

WORKFLOW_LIVE_DISCOURSE and workspace/account policies govern execution. Uploads, user API keys, administrative account management, token lifecycle and triggers are outside this port. Local HTTP is simulated. Real site permissions, review queues, notifications, private-message access and persistence, hosted Review Help and organization persistence remain unverified without a suitable Discourse account and Review deployment.

Copper workflow nodes

Copper provides 32 native CRM operations for people, companies, opportunities, leads, projects, tasks, activities and supporting account/list metadata. Save a copperApi account with Copper API key and Copper API key owner email, or Advanced JSON {"apiKey":"YOUR_KEY","email":"owner@example.com"}. Requests use the fixed https://api.copper.com/developer_api/v1 API and the saved key owner's email. Nodes cannot replace the acting user, authorization headers or origin. These Developer API keys differ from OAuth tokens; OAuth issuance and renewal are not implemented.

Test connection reads /developer_api/v1/users/me, verifies a safe user ID and matches the saved key-owner email. Resource permissions still apply to every operation. Keep record IDs as exact positive decimal text; activity type ID 0 is supported. IDs sent in JSON bodies follow Copper's numeric identifier contract and must fit a JavaScript safe integer; unsupported larger body IDs are rejected before sending, while path IDs remain exact text. Create/update operations expose resource-specific fields. Updates need at least one changed field; false, zero and supported empty values remain explicit. Blank optional text is omitted when saved; to send a supported empty value, use an expression that renders an empty string. Only user activities can be created or modified. A delete succeeds only when Copper confirms the requested ID and is_deleted: true. A later activity read can return a deletion stub with that indicator instead of the former activity details. Unsupported raw activity history objects are omitted from the typed output.

Search and list operations use one page, with a local maximum of 200 records and Copper's first-100,000-record search boundary. Filters depend on the chosen resource. Follow the explicit next page yourself; there is no automatic pagination or write retry. CRM writes can change assignments and activity history or invoke configured automations. Inspect Copper before repeating an uncertain request.

WORKFLOW_LIVE_COPPER and workspace/account policies govern execution. File transfers, bulk jobs, arbitrary HTTP, OAuth lifecycle and trigger execution are outside this port. Local HTTP is simulated. Real record/field permissions, automation effects, deletions and persistence, hosted Review Help and organization persistence remain unverified without a Copper account and Review deployment.

Netlify workflow nodes

Netlify provides 31 native operations for current-user/site metadata, deploy/build controls, build-hook definitions, forms/submissions, read-only DNS metadata and deployed branches. Save a netlifyApi account with Netlify access token, or Advanced JSON {"accessToken":"YOUR_TOKEN"}. Use an existing personal or OAuth access token against the fixed https://api.netlify.com/api/v1 API. Test connection reads /api/v1/user and verifies a user identity. The token still needs access to the chosen team, site and operation. Nodes cannot change the host or token; issuance and automatic renewal are not provided. Protocol details follow the Netlify API reference.

Use exact text resource IDs. Each list performs one request with a conservative local limit of 100 items; where exposed, page starts at 1 and per_page defaults to 50. There is no automatic Link following. Get submission preserves the array response documented by the current API. Returned metadata omits site passwords/environment settings, deploy access tokens and build-hook trigger URLs. Read outputs are selected metadata rather than complete provider objects or arbitrary submission fields.

Site renaming, build starts, deploy cancellation/restoration/locking, hook changes and deletions are writes. They can consume build capacity, change the public site, invoke configured notifications or permanently delete resources. Starting a build does not mean the build finished; read its status explicitly. Build and deploy reads preserve a failed resource state and its declared error text so the workflow can inspect it. Hook definition actions do not install a workflow trigger. Writes are not retried automatically; inspect Netlify before repeating uncertain work.

WORKFLOW_LIVE_NETLIFY and workspace/account policies govern execution. Deploy uploads, binary transfers, environment secrets, domain purchases, billing, OAuth lifecycle and triggers are outside this port. Local HTTP is simulated. Real site/team authority, builds, published-site effects, notifications and deletion persistence, hosted Review Help and organization persistence remain unverified without a Netlify account and Review deployment.

DeepL workflow nodes

DeepL provides three actions: translate an ordered array of plain texts, list text-translation languages, and read character usage. Save a workspace deepLApi account with DeepL API key and an explicit DeepL API plan (free or pro), or Advanced JSON {"apiKey":"YOUR_TRANSLATION_KEY","apiPlan":"free"}. The plan fixes api-free.deepl.com or api.deepl.com; the key suffix does not choose it. Admin keys are rejected. Test connection reads usage only and accepts zero counts and the unlimited-plan sentinel; it does not identify a human, prove translation permission or promise available quota. No key issuance, renewal or plan conversion occurs.

Translate accepts 1–50 separate nonempty texts and a target language. Order and duplicates are preserved; texts are not trimmed, joined or split automatically. Optional source language, sentence splitting, formatting, formality and model preferences are explicit typed fields. Context is sensitive surrounding text sent to DeepL, not instructions: leaving the control untouched omits it; editing then clearing it saves an explicit empty string. False and the splitting value 0 retain their meaning. The local serialized body limit is 120 KiB, with a separate 8 KiB header budget; reduce the submitted array if a limit is reached. Context counts toward request size even when not billed. Translation consumes provider quota and can incur charges; a model preference is not a guarantee. After an uncertain response, inspect usage before repeating. See the translation API and usage API.

Language reads use /v3/languages?resource=translate_text and preserve source/target eligibility, feature status and beta availability. They are response snapshots, not guarantees that every submitted language pair or option is available to the selected key. WORKFLOW_LIVE_DEEPL and workspace policies apply. Documents, glossaries, HTML/XML translation, administration, arbitrary requests, automatic retries and n8n parameter conversion are outside this profile.

Cloudflare workflow nodes

Cloudflare provides six actions: read the saved zone, list/get/create/update/delete DNS records. Save a workspace cloudflareApi account with Cloudflare API token, Cloudflare account ID, Cloudflare zone ID and Cloudflare zone name, or Advanced JSON {"apiToken":"YOUR_TOKEN","accountId":"YOUR_LOWERCASE_HEX_ACCOUNT_ID","zoneId":"YOUR_LOWERCASE_HEX_ZONE_ID","zoneName":"example.com"}. IDs must be exactly 32 lowercase hexadecimal characters; the zone name is a public lowercase ASCII/Punycode domain without a URL or trailing dot. Every DNS action first verifies the saved zone, name and account at fixed api.cloudflare.com. Test connection reads that zone and one A-record page. It proves those reads only, without identity or write-permission certification.

This finite profile supports A, AAAA and CNAME records. Supply a full owner name within the saved zone, content matching its type, explicit TTL (1 for Auto or 60–86400 seconds), and explicit proxy choice. Proxy on requires Auto TTL. A private address is allowed as DNS data; it does not become an HTTP destination. Create/update/delete can change public routing or cause outages. Update and delete read the selected record first. Update requires its ID and the complete selected type/name/content/TTL/proxy tuple, preserves type and name, and sends only those five explicit fields by PATCH; at least one content, TTL or proxy value must change. No fields are borrowed from defaults. The preflight is not a transaction or lock. Delete requires a matching returned ID for confirmed deletion; a response without that acknowledgement leaves the outcome unknown. Do not automatically repeat uncertain writes.

Lists use one explicit bounded page; inspect continuation and completeness metadata. WORKFLOW_LIVE_CLOUDFLARE and workspace policies apply. This adds native DNS actions to the existing Cloudflare card. The shipped n8n Cloudflare node's certificate actions are not implemented, and there is no import conversion or full-node parity. Other record types, renaming, kind conversion, Workers, security administration, certificate lifecycle, automatic pagination/retries and DNS propagation or TLS guarantees are outside this profile. See the DNS records reference.

BambooHR workflow nodes

BambooHR provides three reads: employee-field metadata, the published employee directory, and one employee. Save a workspace bambooHrApi account with BambooHR company subdomain and BambooHR API key, or Advanced JSON {"subdomain":"your-company","apiKey":"YOUR_40_CHARACTER_HEXADECIMAL_KEY"}. The lowercase company label fixes https://your-company.bamboohr.com/api/v1; no legacy gateway fallback occurs. The key is preserved exactly and uses Basic authentication. Test connection reads employee 0 without a field query. A returned string ID 0 is valid for an unlinked principal; success does not prove a linked human, administrative rights or permission to read other employees.

Employee IDs are exact decimal strings, with no leading zero except 0; use 0 only to read the authenticated principal. Employee reads default to firstName, lastName, jobTitle, department and workEmail. Choose only the ten supported directory fields; an explicit empty [] requests ID only. Missing, null and empty fields remain distinct, and unreturnedFields identifies requested values that were not returned. A supervisor name is not an employee ID. Confidential HR data can enter workflow history; outputs omit photos, SSNs, payroll and custom fields.

The directory is the company's published subset, not a complete employee roster. Metadata/directory reads validate the entire bounded response before applying the local result limit. Inspect truncation and completeness metadata. A 403 may mean insufficient access rather than an invalid key; directory 404 is an endpoint failure, not an empty roster, while an employee 404 means not found. WORKFLOW_LIVE_BAMBOOHR and workspace policies apply. Writes, reports, time off, payroll, binaries, key lifecycle, automatic pagination/retries and n8n parameter conversion are outside this profile. See BambooHR authentication and the API reference.

These three profiles use selected workspace accounts and reject redirects. Local provider responses are simulated; live provider acceptance, quota/billing effects, DNS persistence/propagation, real HR permissions, hosted Review Help, organization scope, durable storage and Linux behavior remain unverified without the matching credentials and environment.

Mailgun workflow nodes

Mailgun provides six operations: send a message or named template, read the saved domain, list or read domain templates, and search a finite event-log window. Save a workspace mailgunApi account with Mailgun API key, Mailgun region (us or eu) and Mailgun sending domain, or Advanced JSON {"apiKey":"YOUR_KEY","region":"us","domain":"mg.example.com"}. The saved region and domain fix all destinations. These are existing manual keys; no key or OAuth lifecycle is provided. See the official message API and domain API.

Test connection reads the selected domain. A valid sending-only key may fail that read; failure does not prove the key cannot send, and success does not certify domain readiness, write permissions or delivery. Save the required fields explicitly; the connection check does not issue or expand permissions. Sending targets one explicit email address, can incur charges and may be forwarded downstream. Successful sending means queued acceptance, never delivery. Select text/HTML or an existing domain template; template variables keep supported string, number and boolean values. False for Generate text leaves the provider default rather than explicitly requesting a text alternative. Templates expose bounded metadata, not their source or secret headers.

To send an intentionally empty subject, map Subject from a workflow input such as {{input.subject}} whose runtime value is empty. A blank fixed field is omitted by the workflow editor.

Event searches use the current Logs API, a mandatory saved-domain filter and a finite duration/limit; a POST search is a read. Inspect page/completeness metadata. Account-wide tag reads, attachments, bulk recipients, cc/bcc, scheduling, test mode, administration, arbitrary requests, automatic pagination and retries are outside this profile. If a send response is lost or invalid, inspect Mailgun before repeating. WORKFLOW_LIVE_MAILGUN and workspace policies apply.

Microsoft Excel workflow nodes

Microsoft Excel adds 13 native operations for worksheets, tables, rows and explicit cell ranges. Save a workspace microsoftExcelOAuth2Api account using Microsoft Excel access token and Business OneDrive workbook item ID, or Advanced JSON {"accessToken":"EXISTING_DELEGATED_TOKEN","workbookItemId":"YOUR_ITEM_ID"}. This is a manual existing delegated work/school token and one saved .xlsx workbook in business OneDrive. All operations use that workbook; block inputs cannot substitute another account or workbook. Existing entries missing these fields need an explicit update before using the native actions. No automatic credential conversion, OAuth setup or renewal occurs.

Test connection reads at most one table in that workbook and accepts an empty typed list. It needs the documented delegated Files.Read access; most other Excel operations, including several reads, document Files.ReadWrite. The check does not prove identity or write permission. Personal/consumer accounts, application-only grants, other drives/SharePoint routes and workbook formats are outside this profile. See table discovery and permissions and Excel storage/session behavior.

Use returned worksheet/table IDs. Row updates and deletion take a positional index, which can change during coauthoring. Append/update/delete require the explicit ordered column ID/index/name schema; one same-table read checks it before writing. A schema check is not a transaction or lock. Worksheet/table discovery reads one finite page; row/column reads support explicit manual offsets. Inspect incomplete/unsupported-continuation metadata instead of assuming all data was returned.

Ranges require exact uppercase A1 rectangles and rectangular matrices, at most 100 rows, 50 columns and 5,000 cells. False, zero and valid empty text keep their meaning. Null cells in updates mean leave unchanged; append does not accept null. Empty text clears a value. Ambiguous formula-like text, formula fields, broadcasting and arbitrary JSON are rejected. Numeric negatives remain supported as numbers. Sessionless writes persist and can recalculate existing formulas, propagate table formulas or affect references. Clear contents removes values and formulas; worksheet/row deletion is destructive. The clear endpoint accepts empty 200 or 204 as a bounded interpretation of contradictory official prose/examples; other operation statuses remain specific. No sessions, async jobs/polling, scripts/macros, formatting, transactions, automatic splitting or retries are provided. After an uncertain write, reconcile the workbook before repeating. WORKFLOW_LIVE_MICROSOFT_EXCEL and workspace policies apply.

OpenWeatherMap workflow nodes

OpenWeatherMap provides four bounded reads: current weather, a five-day forecast in three-hour intervals, direct geocoding and reverse geocoding. Save a workspace openWeatherMapApi account with OpenWeatherMap API key, or Advanced JSON {"apiKey":"YOUR_KEY"}. Requests use fixed api.openweathermap.org; the key is excluded from results and errors. Test connection reads current weather at fixed London coordinates. It does not identify the account, prove access to other products or certify fresh weather conditions. Location reads disclose the requested location to OpenWeather and consume quota. No One Call subscription or paid access is created. See current weather, forecast and geocoding.

Use finite numeric latitude/longitude, including zero and negative values. Units default to metric; standard uses kelvin and imperial uses Fahrenheit. Forecast timestamps, sunrise/sunset and timezone offsets use seconds; the fetch time is separate from the provider's measurement/forecast time. Forecast precipitation uses three-hour volumes, unlike current one-hour values. Counts are finite; short, repeated, irregular or mismatched timestamp sets retain valid rows with explicit incomplete metadata. A full requested count does not certify an entire five-day window or daily summaries.

Geocoding returns multiple candidates for ambiguous names and never automatically chooses one or fetches its weather. Empty candidates mean no result, not authenticated account identity or exhaustive search. HTTPS geocoding and strict 200 acceptance are selected consumer policies; published error/status conventions and live TLS behavior have not been fully certified. One Call, history, air pollution, icon downloads, callbacks, arbitrary queries, automatic pagination and retries are excluded. WORKFLOW_LIVE_OPENWEATHERMAP and workspace policies apply.

The three profiles use selected workspace credentials and reject redirects. Their local HTTP tests are simulated. Actual provider acceptance/effects, hosted Review Help, organization paths, durable storage and Linux behavior remain unverified without the corresponding credentials and environment.

Pushover workflow nodes

Pushover provides three operations: send a message, validate a recipient and list sounds. Save a workspace pushoverApi account with Pushover application key, or Advanced JSON {"apiKey":"YOUR_30_CHARACTER_APP_KEY"}. The key belongs to the application; each node requires its own explicit recipient key. Requests use fixed api.pushover.net. Test connection reads application credits; it does not prove recipient access, identity, delivery or general permissions. See the official API and credit endpoint.

A recipient can be a group and notify multiple people. An invalid device can fall back to all active devices. Priority defaults to normal (0); high priority (1) can bypass quiet hours. Successful sending means queued acceptance, never confirmed delivery. Optional title, device, sound and supplementary HTTPS link are typed fields; a link title requires a link. Emergency priority, attachments, HTML, receipts, cancellation, token lifecycle, automatic pagination and retries are unsupported. The application key is sent in documented form bodies or query parameters and is excluded from outputs and errors. WORKFLOW_LIVE_PUSHOVER and workspace policies apply.

Agile CRM workflow nodes

Agile CRM provides 18 operations for contacts, companies, deals, notes, tasks and event reads. Save a workspace agileCrmApi account with Agile CRM account email, Agile CRM API key and Agile CRM subdomain, or Advanced JSON {"email":"operator@example.com","apiKey":"YOUR_KEY","subdomain":"your-account"}. The lowercase subdomain fixes https://your-account.agilecrm.com; the saved email and key form Basic authentication. The key is limited to 4,096 printable ASCII characters. Test connection reads one contact page and accepts its documented empty 204 result. It does not certify identity or write permission. See the provider API reference.

Use exact decimal text IDs when values exceed JavaScript's safe integer range. Times use each operation's documented epoch seconds. Contact and company property updates first read and check the record kind, then issue one update with the same account. Company lists use their documented form request; contact lists use query parameters. Inspect page cursors and completeness metadata; unpaginated reads do not promise all records. Contact/company reads can return not found through 204; that status is not accepted for other operations without documentation. The task-read shape and note route use explicit bounded compatibility interpretations described in the integration study. CRM writes may trigger account automation. Custom fields, arbitrary filters, deletion, campaigns, administration, OAuth, automatic pagination and retries are unsupported. WORKFLOW_LIVE_AGILECRM and workspace policies apply.

Grafana workflow nodes

Grafana provides 16 operations for folders, teams, team members, annotations, data source metadata, search and dashboard summaries. Save a workspace grafanaApi account with Grafana installation URL and Grafana Bearer token, or Advanced JSON {"baseUrl":"https://metrics.example.com/grafana","apiKey":"YOUR_EXISTING_TOKEN"}. The saved public HTTPS installation on port 443 and optional subfolder fix all destinations. Use an existing service-account token or a legacy token the installation still accepts. This is a legacy /api compatibility profile: Grafana documents deprecation beginning with version 13 while retaining those routes. There is no automatic /apis fallback or claim of compatibility with every release or edition. See the official API reference and team permissions.

Test connection requires team-search read permission and validates a small typed result. A folder-only token can fail this check; success does not prove token identity under every anonymous-access configuration or certify other permissions. Team member lists first validate the readable team. Team and membership writes change access; team deletion removes the selected team, and annotation creation persists a point or region using epoch milliseconds. Data source results exclude URLs, authentication settings and secret configuration. Dashboard summaries use search metadata and never expose raw dashboard JSON. Review page and truncation metadata: capped or unpaginated results do not certify global completeness. Dashboard writes, queries/SQL, data source writes, user/organization administration, token lifecycle, OAuth and retries are unsupported. WORKFLOW_LIVE_GRAFANA and workspace policies apply.

The three adapters reject redirects and unexpected statuses, use selected workspace accounts, and expose typed projections. Their HTTP tests are simulated. Live-provider effects, hosted Review Help, organization paths and durable persistence remain unverified because suitable credentials and a Review environment are unavailable.

PagerDuty workflow nodes

PagerDuty provides 26 operations for incidents, notes, alerts, logs, services, escalation policies, schedules, on-call records, priorities, users, teams and business services. Save a workspace pagerDutyApi account with PagerDuty API token and optional PagerDuty From email, or Advanced JSON {"apiToken":"YOUR_TOKEN","fromEmail":"operator@example.com"}. Use an existing REST API token, not an Events routing key. Requests use fixed api.pagerduty.com and REST v2 headers. See the official REST schema and authentication guide.

Test connection reads at most one service; an empty services array is valid scoped read access. This does not identify a token owner or prove incident-write authority. Every exposed write requires the valid From email saved in that account; block inputs cannot change the acting user. Creating incidents can notify responders. Incident updates explicitly choose changes such as acknowledgment, resolution, reassignment or reopening; alert updates can resolve, reopen or move an alert. Read the selected operation's effects before running.

Keep IDs as text. Lists perform one bounded request with the supported offset/filter controls; returned links are not followed. Updates require a meaningful change, preserve explicit false/zero/empty values where supported and confirm returned target/state. Returned incident title, urgency, service, priority, escalation policy and assignees must match requested values. Conference-bridge details and incident bodies are conditional API echoes; absence does not verify those changes. Resolution text is added to a resolve log, and escalation-level acceptance does not independently verify the reached level. WORKFLOW_LIVE_PAGERDUTY and workspace policies apply. Events ingestion, token lifecycle, resource administration, attachments, automatic pagination and write retries are outside this port. HTTP is simulated locally; live permissions, responder notifications, hosted Review Help and organization/durable persistence remain unverified.

Sendy workflow nodes

Sendy provides eight operations: list brands and lists, read subscriber status/count, subscribe or update, unsubscribe, delete, and create a campaign draft. Save a workspace sendyApi account with Sendy installation URL and Sendy API key, or Advanced JSON {"baseUrl":"https://mail.example.com/sendy","apiKey":"YOUR_KEY"}. The saved public HTTPS installation and optional subfolder fix request destinations. The API key is sent in documented form bodies; unsubscribe uses its documented email/list form without inventing key authentication. See the official API.

Test connection reads brands. The supported bounded JSON array of id/name records and HTTP 200 acceptance policy are conservative inferences from the documentation's collection description; its exact envelope is undocumented and no live installation was tested. Another installation's valid response may therefore be rejected. Plain-text errors such as No brands found do not prove identity or writes.

Use exact brand/list identifiers. Subscriber upsert can trigger confirmation mail or automation according to list settings; bypassing double opt-in is unsupported. Unsubscribe changes delivery eligibility and deletion removes the subscriber. Campaign creation requires explicit Draft mode and sends send_campaign=0; send/schedule/recipient fields are rejected. Acknowledgment is not independent persistence or delivery proof. Counts preserve zero; supported empty text and false values are forwarded explicitly. WORKFLOW_LIVE_SENDY and workspace policies apply. Campaign sending/scheduling, arbitrary custom fields, automatic pagination and write retries are outside this port. HTTP is simulated; installation compatibility, real mail effects, hosted Review and organization/durable persistence remain unverified.

Raindrop workflow nodes

Raindrop provides 30 operations for bookmarks, collections, tags, highlights, filters and user metadata. Save a workspace raindropOAuth2Api account with Raindrop access token, or Advanced JSON {"accessToken":"YOUR_TOKEN"}. Use an existing test or OAuth token at fixed https://api.raindrop.io/rest/v1; this form does not issue or renew it. Test connection validates the current-user response, not write or sharing permissions. See token setup and the API documentation.

Use exact numeric resource IDs and explicit collection selection. Special collection IDs are accepted only on operations that document them; destructive collection and tag operations require positive collection IDs. Lists make one bounded request with exposed page controls. Outputs project selected metadata and omit raw capabilities and page content.

Bookmark creation can make Raindrop fetch the supplied page. Collection publication changes sharing, deletions affect saved data or trash, and tag merging changes matching bookmarks. Merge source tag names must be distinct. Updates need a changed field; supported false, zero, empty text and empty tag arrays retain their meaning. Highlight updates support note/color, not text replacement. WORKFLOW_LIVE_RAINDROP and workspace policies apply. Binary transfers, token lifecycle, ambiguous bulk update/delete, automatic pagination and write retries are outside this port. HTTP is simulated; real fetch/sharing/deletion effects, hosted Review Help and organization/durable persistence remain unverified.

Pushbullet workflow nodes

Pushbullet provides 15 native operations for the current user, devices, chats, note/link pushes, channel subscriptions and channel information. Save a workspace pushbulletOAuth2Api account with Pushbullet access token, or Advanced JSON {"accessToken":"YOUR_TOKEN"}. Use an existing personal or OAuth token. Requests use the fixed https://api.pushbullet.com/v2 API and Access-Token header; token issuance and renewal are not provided. Test connection reads /v2/users/me and checks its user identifier. Optional empty name/email fields may be absent. This does not certify write permissions or quota. See the Pushbullet API reference.

Keep identifiers and channel tags as text, and timestamps as decimal epoch seconds. Lists perform one bounded request and expose a cursor for explicit continuation. Pushbullet can omit empty arrays and false/empty fields; the adapter accepts documented omissions and typed deleted-object stubs. Deleted metadata does not confirm a successful create/update. Create push requires note or link and an explicit recipient choice: self broadcasts to the account's devices, device targets one device, email can send fallback email, and channel broadcasts to its subscribers. Notes need meaningful title or body; links require an HTTP(S) URL. The link is message content and is not fetched by Confluye. Supply only the selected recipient's target field.

Chat and subscription updates require explicit muted true/false; push updates require an explicit dismissed choice. Creating/deleting chats or subscriptions changes this account's relationships. Pushes consume the provider's allowance and may notify people; API acceptance is not delivery confirmation. WORKFLOW_LIVE_PUSHBULLET and workspace/account policies govern execution. Files, SMS, device provisioning, channel creation, delete-all, OAuth lifecycle, triggers, automatic pagination and write retries are outside this port. Local HTTP is simulated. Real token access, quotas, notifications, recipient delivery and persistence, hosted Review Help and organization persistence remain unverified without a suitable account and Review deployment.

UptimeRobot workflow nodes

UptimeRobot provides 24 native API v3 operations for current-user metadata, monitors, personal Email alert contacts, maintenance windows and public status pages. Save a workspace uptimeRobotApi account with UptimeRobot API v3 token, or Advanced JSON {"apiKey":"YOUR_TOKEN"}. Enter an existing v3 token without the Bearer prefix. Requests use Bearer authentication on fixed https://api.uptimerobot.com/v3. The current official OpenAPI establishes this contract; old Basic/v2 prose on the documentation page is not the v3 authentication contract. Legacy body-key v2 authentication, token issuance and renewal are unsupported.

Test connection reads /v3/user/me and validates the direct current-user response. It does not prove write permissions, account-wide access or compatibility of old monitor-specific keys. The selected token's permissions and plan remain authoritative. Resource IDs use safe JSON integers. Lists return one page with a maximum of 200 entries and an explicit cursor: maintenance windows use text, while monitors, alert contacts and status pages use numbers. Returned pagination URLs are not followed. Output omits monitor URLs, monitored credentials, callback/device values and other capability-bearing fields. Reads expose selected metadata rather than complete provider objects.

Monitor creation supports HTTP monitors only, with explicit name, monitored URL, interval and timeout. Monitored URLs are content sent to UptimeRobot, not destinations fetched by the Confluye gateway; embedded credentials are rejected. Monitor alert assignments require contact ID, threshold and recurrence. Writes can change checks or send alerts. Maintenance creation supports one dated window, not recurring schedules; date and time are sent as entered and duration is in minutes. It suppresses alerts for the selected monitors; auto-add can include all monitors. Updates require a changed field and preserve supported false, zero and empty arrays.

Status-page creation requires an explicit ENABLED or PAUSED choice. Publication exposes monitoring availability. Supplied monitor selections replace the list; an empty list detaches all monitors, while [0] alone selects every monitor. Do not combine that sentinel with other IDs. Auto-add can expose existing and future monitors. Email contact creation may require provider verification. WORKFLOW_LIVE_UPTIMEROBOT and workspace/account policies govern execution. Monitor credentials/custom HTTP payloads, non-HTTP creation, recurring maintenance, binaries, administrative/token lifecycle, workflow triggers, automatic pagination and write retries are outside this port. Local HTTP is simulated. Real plan/key authority, alert verification/delivery, suppression timing, public publication and persistence, hosted Review Help and organization persistence remain unverified without an account and Review deployment.

Sentry workflow nodes

Sentry provides 11 native operations for accessible organizations, projects, teams, issues, selected event metadata and issue state updates. Save a workspace sentryIoApi account with Sentry API token, or Advanced JSON {"token":"YOUR_TOKEN"}. Use an existing token against fixed global https://sentry.io/api/0 with Bearer authentication. Regional/self-hosted hosts, OAuth issuance and renewal are unsupported; the reference's OAuth/server credential modes are not aliases for this account. See Sentry authentication and its official public API schema.

Test connection reads one accessible-organizations page. An empty array is valid scoped read access, not an identity or write check. Organization/project/team reads require the applicable org:read, project:read or team:read grants; issue/event reads require an applicable event-read grant, and issue changes need event-write authority. Sentry accepts some broader grant alternatives. These are server-enforced permissions; Confluye does not issue grants or infer them from a successful connection check.

Keep IDs/slugs and event counts as exact text. Lists read one bounded page of at most 100 items; continue with the returned cursor, including its punctuation/signs, not a URL. Issue lists use the current organization endpoint with a supported project filter and explicit all/unresolved/resolved/ignored choice; all sends an empty query. Event reads return selected metadata and omit raw event bodies, source content and access capabilities. Permission-dependent and nullable fields retain their declared shapes.

Issue updates require at least one supported change: status, priority, seen, bookmarked or subscribed. False values remain explicit. Updates can change triage and notification subscriptions; uncertain writes are not retried. Inbox review-state mutation is excluded because its complete supported confirmation contract was not established. The current update prose permits partial changes despite a conflicting generated required-field list; this adapter exposes only the documented bounded patch. WORKFLOW_LIVE_SENTRY and workspace/account policies govern execution. Destructive organization/project changes, membership administration, keys/DSNs, ingestion, attachments, raw events, arbitrary searches, OAuth lifecycle, triggers and automatic pagination are outside this port. Local HTTP is simulated. Real grants, region compatibility, subscription effects and persistence, hosted Review Help and organization persistence remain unverified without an account and Review deployment.

GetResponse workflow nodes

These workflow nodes use a saved workspace account. Personal agent and browser credentials are separate connections.

GetResponse provides 19 native operations for account identity, contacts, campaigns (contact lists), custom fields and tags. Save a getResponseApi account with GetResponse API key, or Advanced JSON {"apiKey":"YOUR_KEY"}. Supply the existing key without the api-key prefix. Requests use the fixed SMB https://api.getresponse.com/v3 API and X-Auth-Token authentication. MAX/custom domains, token issuance and renewal are unsupported. Test connection reads /v3/accounts and checks account ID and email; it does not certify write permissions. See the GetResponse API reference.

Keep IDs as text. Lists return one page with a local maximum of 100 entries; continue explicitly. Create contact requires the target campaign and email. HTTP 202 returns accepted: true, completed: false, status: queued: the contact has not been confirmed as stored or subscribed. Processing may reject it or require double opt-in. Inspect GetResponse before repeating an uncertain request. Updates require at least one field; supplied tag and custom-field arrays replace existing values, and an empty array clears them. Moving a contact to another campaign changes its history. Campaign operations manage contact lists, not newsletter sending. Custom-field creation requires name, type, format, hidden and values; radio and select formats require at least two choices. hidden and isDefault are text "true"/"false", not JSON booleans.

Writes may change subscriptions, trigger automations or delete data. WORKFLOW_LIVE_GETRESPONSE and workspace/account policies govern execution. Automatic all-page traversal, write retries, binary transfers and the n8n GetResponse trigger are outside this port. Local HTTP is simulated. Real account permissions, queued-contact completion, consent/subscription effects, automation, hosted Review Help and organization persistence remain unverified without a suitable GetResponse account and Review deployment.

Freshservice workflow nodes

These workflow nodes use a saved workspace account. Personal agent and browser credentials are separate connections.

Freshservice provides 20 native operations for tickets, conversations, ticket tasks, requesters and supporting field/workspace/agent/group metadata. Metadata resource variants count as one operation. Save a freshserviceApi account with Freshservice tenant subdomain and Freshservice API key, or Advanced JSON {"domain":"my-company","apiKey":"YOUR_KEY"}. Enter only the lowercase DNS label before .freshservice.com, without URL, dot, path or port. The fixed saved tenant uses /api/v2 with API-key Basic authentication. Nodes cannot change its tenant or acting account. See the Freshservice API v2 reference.

Test connection reads one ticket page. An empty collection establishes scoped read access, not identity or write authority. The API-key agent still needs the relevant resource permissions. Ticket creation requires exactly one requester ID or email. Use exact text path IDs; JSON-body IDs must fit a positive safe integer. Updates require a changed field. Supplied tags replace the existing array; false, zero and supported empty values remain explicit. Requester updates must explicitly choose can_see_all_tickets_from_associated_departments, because the provider otherwise defaults it to false and can change visibility. Ticket updates cannot move a ticket to another workspace. Workspace selectors on supported reads/creates select Freshservice resources and do not bypass Confluye workspace permissions.

Notes require an explicit private flag. Public notes, replies and notification recipients can send messages. Only notes support conversation editing. Lists return one bounded page of at most 100 records, without automatic traversal. WORKFLOW_LIVE_FRESHSERVICE and workspace/account policies govern execution. Attachments, assets, account administration, token lifecycle, triggers and automatic write retries are outside this port. Local HTTP is simulated. Real agent/workspace permissions, notifications, requester visibility and write persistence, hosted Review Help and organization persistence remain unverified without a Freshservice account and Review deployment.

Lemlist workflow nodes

These workflow nodes use a saved workspace account. Personal agent and browser credentials are separate connections.

Lemlist provides 13 native operations for team/sender metadata, campaigns, statistics, leads, activities and pausing. Save a lemlistApi account with Lemlist API key, or Advanced JSON {"apiKey":"YOUR_KEY"}. The existing key uses Basic authentication on fixed https://api.lemlist.com/api. Test connection reads /api/team?version=v2 and validates team ID and name. It does not establish campaign permissions or write authority. Returned team metadata omits webhook capabilities. See the Lemlist API reference.

Create lead can start outreach. Adding a lead to an auto-launch campaign may send messages. The required acknowledgeAutoLaunch field must explicitly be true; it has no default and is not sent to Lemlist. Disabling enrichment does not prevent sending, and this adapter offers no atomic paused creation. Confirm the campaign's settings before running. Lead updates can also change shared contact details. Supported empty strings clear fields; updates require at least one changed field. Get lead requires exactly one lead ID or email and returns the v2 array. Lead pause always requires a campaign; campaign pause does not guarantee that already scheduled or in-flight work stops.

Lists make one request: campaigns allow at most 100 entries, campaign leads 500 and activities 100. Continue using the declared page or offset; there is no automatic traversal or rate-limit retry. WORKFLOW_LIVE_LEMLIST and workspace/account policies govern execution. Start/resume operations, explicit send operations, enrichment, global unsubscribe, billing, OAuth lifecycle and the n8n Lemlist trigger are outside this port. Local HTTP is simulated. Real team/campaign permissions, auto-launch outreach, shared-contact updates, pause effects and persistence, hosted Review Help and organization persistence remain unverified without a Lemlist account and Review deployment.

Grist workflow nodes

Grist provides nine native operations: authenticated profile, document, table and column metadata, record reads and create/update/upsert/delete. Save a gristApi account with Grist site URL and Grist API key, or Advanced JSON {"baseUrl":"https://docs.getgrist.com","apiKey":"YOUR_KEY"}. This supports a public HTTPS root origin on port 443; private hosts and subfolder installations are unsupported. The saved origin and key stay with the selected account. Test connection reads /api/profile/user and rejects anonymous profiles, including anonymous HTTP 200 responses. Document permissions still apply. Protocol behavior follows the Grist API reference; deployed versions can differ.

Use exact document/table IDs and column IDs. Record IDs must fit a positive JavaScript safe integer. The update collection accepts canonical decimal text from the editor and sends a JSON number without rounding; signs, whitespace, exponent text and larger integers are rejected. Dynamic column maps use JSON editors. Grist cell values retain false, zero, empty text, null, decimals and tagged arrays; date/datetime tags use epoch seconds. noparse defaults to true to preserve strings literally; false asks Grist to parse them by column type. Upsert requires nonempty matching criteria; review whether additions and updates are enabled before executing.

Record reads return at most 500 rows in one request. Metadata lists fail above the local 1,000-item bound. No offset, cursor or automatic all-record traversal is invented. Writes accept at most 100 records and may trigger document formulas, webhooks or automation; deletion is permanent. Current upsert result metadata is checked explicitly, so a server version with a different response contract can fail after applying a write. Writes are never retried automatically: inspect Grist before repeating an uncertain request. WORKFLOW_LIVE_GRIST and workspace/account policies govern execution. Binary files, document administration, schema writes, OAuth lifecycle and workflow triggers are outside this port. Local HTTP is simulated. Real account/document permissions, formulas, write persistence, hosted Review Help and organization persistence remain unverified without a suitable Grist account and Review deployment.

Harvest workflow nodes

Harvest provides 19 native operations, with resource selectors for record reads/deletes, resource-specific create/update fields for clients, contacts, projects, tasks, time entries and expenses, timer restart/stop, company settings and the current user. Resource variants are not counted as separate operations. Save a harvestApi account with Harvest access token, Harvest account ID and Harvest application identification, or Advanced JSON {"accessToken":"YOUR_TOKEN","accountId":"123456","userAgent":"My automation (ops@example.com)"}. Use an existing personal or OAuth access token. The account ID is exact positive decimal text, fixed with the credential. Supported application identification is printable ASCII, at most 256 characters, containing an application name and parenthesized contact email or HTTP(S) URL. Harvest requires application identification; malformed or missing values can prevent access. Token issuance and automatic renewal are not provided.

Requests use fixed https://api.harvestapp.com/v2, with the saved account and application headers. Test connection reads /v2/users/me and validates a user ID and email. It does not establish project, expense or administrative authority. Use exact positive text IDs; IDs in JSON write bodies must fit a safe integer and are converted without rounding. Dates use real YYYY-MM-DD calendar dates. Rates, costs and hours preserve decimals without currency conversion. Duration and timestamp time entries require the matching company mode; omitting end information can start a timer. Update operations require at least one change. False, zero and supported empty values remain explicit; blank optional text is omitted by the editor, so use an expression to send a supported empty string.

Lists perform one request with a local maximum of 100 items. Choose page or the returned cursor, never both; provider pagination support varies by resource. Expense updates check whether the returned project/category matches the requested reassignment, since locked expenses may ignore those changes. Timer results are checked for the requested running/stopped state. Financial states such as a declined estimate remain inspectable business data. Writes can alter billing records, expenses, assignments and time tracking; do not repeat an uncertain write without inspecting Harvest. WORKFLOW_LIVE_HARVEST and workspace/account policies govern execution. Invoice/estimate writes, payment collection, binary receipts, administrative changes, OAuth lifecycle and triggers are outside this port. Local HTTP is simulated. Real account roles, timer mode, locked-expense behavior, billing effects, write persistence, hosted Review Help and organization persistence remain unverified without a Harvest account and Review deployment.

Microsoft To Do workflow nodes

Microsoft To Do provides 20 native operations for task lists, tasks, checklist items and linked resources. Save a microsoftToDoOAuth2Api account using Microsoft To Do access token, or Advanced JSON {"accessToken":"YOUR_TOKEN"}. Use an existing delegated Microsoft Graph token for the global cloud. Tasks.Read or Tasks.ReadWrite permits reads; writes require Tasks.ReadWrite, authority over the selected resources and explicit request confirmation. The fixed API is https://graph.microsoft.com/v1.0/me/todo; nodes cannot replace the acting user or cloud. Token issuance, automatic renewal and application-only tokens are unsupported.

Test connection reads /v1.0/me/todo/lists?$top=1 and validates a task-list collection. An empty collection is valid scoped read access; it does not prove identity or write permission. Use exact opaque text IDs, including their punctuation. Lists perform one request, defaulting to 50 items with a local maximum of 100. To continue, supply the complete returned nextLink for the same resource. Its $top controls the continued page size; the separate Top field is ignored when nextLink is supplied. The adapter restricts it to the fixed Graph origin and exact resource path; it never follows arbitrary URLs or traverses pages automatically. This preserves Microsoft's paging contract.

Task dates retain their local date/time and supported Windows or IANA timezone without automatic UTC conversion. Recurrence validates fields for the chosen pattern and range; enabling a reminder requires its date/time and may notify the user. Task updates support HTML bodies; create supports its declared text/HTML choices. Updates require a changed field and preserve supported false, zero and empty values; use an expression for an optional empty string, because a blank optional text control is omitted on save. Linked resources use declared HTTP(S) URLs. Creation, updates and deletion affect the user's tasks and shared lists; deletions are permanent and writes are not retried automatically. WORKFLOW_LIVE_MICROSOFTTODO and workspace/account policies govern execution. Attachments, national clouds, arbitrary-user access, OAuth lifecycle and webhook triggers are outside this port. Local HTTP is simulated. Real delegated scopes, sharing permissions, timezone acceptance, recurrence/reminder effects, write persistence, hosted Review Help and organization persistence remain unverified without a suitable Microsoft account and Review deployment.

The n8n v1 import profiles cover task lists, tasks and linked resources within explicit parameter limits. Source references never select an account. All twelve native writes, including checklist writes, require confirmation of the actual request; a declined or absent confirmation sends nothing. Provider business extensions are retained in task/list/linked-resource results.

Mailjet workflow nodes

Mailjet provides eight native Email API actions: send an inline or template message, list/get templates, and list/get/create/update contacts. Save mailjetEmailApi with its API key, paired secret key, explicit eu or us architecture and boolean sandboxMode. The form starts visibly in sandbox; Advanced JSON must supply the mode explicitly. The saved account fixes the regional destination. SMS bearer tokens and the historical combined node's SMS/trigger behavior are outside this profile. See the Mailjet API reference.

Test connection reads one template metadata page, accepting an empty page. This proves that read only, not identity, sender verification, send permission, delivery or quota. Sends submit one message with typed sender/recipient objects and at most 50 combined To/Cc/Bcc recipients. Inline sends need text or HTML; template sends need an existing template ID. Ordered inputs, optional empty Subject, priority zero and explicit false retain their meaning. Leaving Subject untouched omits it; editing and clearing saves an empty string. Templates/contacts use canonical positive decimal text IDs within the JavaScript safe-integer limit; larger provider int64 IDs are rejected without rounding. The same ceiling applies to returned message IDs: an oversized ID can fail validation after Mailjet has accepted a send, leaving its effects unknown. Inspect Mailjet before repeating that request. Lists return one explicit bounded page, with no automatic traversal.

Choose the send acknowledgement matching the saved sandbox mode: validation without delivery, or real sending with possible quota effects. Changing the saved mode invalidates the other acknowledgement. Contact writes still change contacts while sandbox is enabled. Empty contact Name and false marketing exclusion remain explicit; changing exclusion does not grant consent or subscribe a person to a list. A successful send response establishes validation or acceptance, never inbox delivery. Timeouts and malformed responses can leave write effects unknown; inspect Mailjet before repeating. Requests are not retried automatically. WORKFLOW_LIVE_MAILJET and workspace policies govern execution; token lifecycle, attachments and automatic n8n conversion are excluded. Provider HTTP is simulated locally; real authentication, delivery, quota effects, Review, organization and durable behavior remain untested.

Demio workflow nodes

Demio provides five native actions: list/get events, get a selected session, register a participant and list session participants. Save demioApi with its existing API key and paired secret. The destination is fixed to https://my.demio.com/api/v1. Test connection uses authenticated GET /ping; it proves acceptance of that pair by ping only, not identity, event access, registration permission or remaining quota. Reads and checks use API allowance. See the current Demio API.

Select positive safe-integer event and session IDs explicitly. Event dates can have null times for on-demand sessions; session detail uses its distinct documented running/time representation. Explicit Active false is retained. Participant filters use the documented hyphenated values while returned statuses contain spaces. Lists make one bounded request and return complete:false; the local limit is not an upstream page parameter or an attendance completeness claim. Report custom fields are typed reads; optional custom registration writes are deferred because current endpoint typing is incomplete.

Registration requires explicit acknowledgement: it grants a sensitive join link and may send configured confirmation, reminder and replay emails or activate integrations; Series settings can retain registration across sessions. The join link is returned as inert sensitive data, never fetched automatically. A failed or uncertain response does not establish rollback or justify an automatic retry. WORKFLOW_LIVE_DEMIO and workspace policies govern execution. Scheduling, cancellation, attendance writes, triggers and token lifecycle are excluded. Provider HTTP is simulated locally; actual permissions, notifications, quotas, Review, organization and durable behavior remain untested.

Customer.io workflow nodes

Customer.io uses two separate native profiles. customerIoTrackApi needs an explicit us/eu region, existing site ID/API key, expected provider environment ID as exact positive decimal text and declared id, email or id_or_email identifier mode. customerIoAppApi needs its own ordinary workspace App key and explicit region; Track keys and cross-workspace service-account send tokens cannot replace it. Enter the environment ID as digits only, within the safe-integer range; Advanced JSON requires a quoted string such as "expectedEnvironmentId":"61". Numeric JSON, signs, leading zeros, fractions and exponents are rejected so formatting cannot round the saved identity. The adapter converts valid text once to compare the numeric provider response. There is no automatic conversion of the historical combined n8n credential. Current regional endpoints follow the Track API and App API, with no legacy beta-path fallback.

Track has six actions: upsert/delete a profile, track a person/anonymous event, and add/remove IDs from an existing manual segment. Its connection check reads /api/v1/accounts/region and matches the returned environment number, region and fixed destination. This does not verify the declared identifier mode or writes. App has five reads: campaign list/get/metrics and segment list/get. Its check lists campaigns and discards resource data; an empty list is valid, but does not identify the provider workspace or certify Track access. Keys and selected local workspace accounts stay separate from operation data.

Use typed flat attribute/event rows and exact person identifiers; reserved identity, subscription and mail-recipient override keys are rejected. Supported false, zero, empty and event null values retain their meaning; profile attribute null is unsupported. Events and audience changes can trigger messages and automations. Segment members require a JSON array of exact IDs; commas within an ID are preserved, and comma-separated text is rejected. Segment mutations use explicit IDs and acknowledgements of an existing manual segment and audience effects; submitted count is not affected count. Profile deletion allows later recreation and does not suppress an identifier. Writes require explicit acknowledgements and return acceptance with effects unverified; no automatic retry follows an uncertain result.

App lists read one bounded response without claiming completeness; archived segments can be absent. Metrics use version 2, UTC and a finite requested window, preserving absent versus empty series and zero values. WORKFLOW_LIVE_CUSTOMERIO_TRACK and WORKFLOW_LIVE_CUSTOMERIO_APP independently govern the profiles. Suppression, identity reassignment, subscriptions, direct transactional sends, webhook lifecycle, arbitrary HTTP and automatic n8n conversion are excluded. Provider HTTP is simulated locally; real key permissions, provider workspace settings, asynchronous audience/message effects, quotas, Review, organization and durable behavior remain untested.

Mattermost workflow nodes

Mattermost provides twelve finite API v4 actions: read the saved user, list that user's teams/channels, get a channel, list channel users, get a user, list/get posts, list reactions, create a post and add/remove the saved user's reaction. Save mattermostApi with baseUrl, existing accessToken and the exact 26-character lowercase expectedUserId. Use a public HTTPS installation on port 443, optionally with a plain subfolder; omit /api/v4, query strings and fragments. The destination and acting user come from the saved account. See the Mattermost API.

Test connection reads /api/v4/users/me and requires the saved ID, username and original integer delete_at:0. Every action repeats that active-user check; the other eleven then make one main request within the same deadline. This local policy rejects missing or nonzero deletion markers. It proves that read at that moment, not every permission, server version or future token validity. IDs remain strings, including when mapped from preceding nodes; numeric whole-expression IDs are rejected before the action request. Post timestamps remain exact decimal text. Lists return only the explicit bounded page.

The three writes require Effects acknowledged to be checked; it starts unchecked. Posts and reactions can persist, notify people or invoke server plugins, previews and audit/retention behavior. That checkbox is not proof of consent. A failed identity check prevents the main action, but a timeout or malformed response after a write can leave its effects unknown. Inspect the server before repeating; there is no automatic retry or rollback. WORKFLOW_LIVE_MATTERMOST and workspace policies govern execution. Uploads, arbitrary endpoints, trigger lifecycle and automatic n8n conversion are excluded. Provider requests are simulated locally; released-server compatibility, live authentication/effects, hosted Review, new-node organization and durable behavior remain untested.

Jotform workflow nodes

Jotform provides seven read actions: user, usage, settings, forms, form questions, form submissions and one submission. Save jotFormApi with an existing apiKey and explicit region: standard, eu or hipaa. The saved region selects the fixed API origin; the key is sent in the APIKEY header. No token belongs in a node input or URL. Selecting the HIPAA origin does not establish account eligibility or compliance. See the Jotform API.

Test connection reads /user and requires HTTP 200, original integer responseCode:200 and a nonempty username. It does not prove form access, key scope, quota or access to another region. IDs are exact positive decimal strings, including large provider IDs; Advanced JSON requires quoted IDs, and mapped numeric IDs reject before the dependent request. Map a returned form ID from {{node.forms.json.data[0].id}} into Form ID, substituting your actual node ID. A list reads one explicit offset/limit page; its next-page hint does not prove exhaustion or a complete total.

Submission answers are omitted unless Include answers is explicitly true. They may contain confidential personal data; unknown profile fields, raw envelopes, private error bodies and URL/IP metadata are omitted. Reads and checks consume API allowance. There are no writes, webhook registration, polling or event triggers: these read actions do not port the reference Jotform Trigger. WORKFLOW_LIVE_JOTFORM and workspace policies govern execution. Local provider HTTP is simulated; live regional responses/permissions/quota, hosted Review, new-node organization and durable behavior remain untested.

Segment workflow nodes

Segment provides six Tracking actions: identify, track, page, screen, group and alias. Save segmentApi with an existing writeKey and explicit us or eu region. This profile uses a printable ASCII key without a colon, with an empty Basic-auth password. The saved region fixes the destination (api.segment.io or events.eu1.segmentapis.com); the destination is not a node field. See the Segment HTTP Tracking API.

Choose Save account. Authentication remains unverified: there is no harmless authenticated Tracking-key read in this profile, so neither saving nor a bulk check sends an event. Public or Config API keys are different credentials. A saved account is sufficient to select for a reviewed workflow; it is not proof that the key or source is valid.

Every action requires an explicit messageId, Delivery choice and unchecked-by-default Effects acknowledged. Use a different message ID for each logical event, preferably mapped from a unique upstream event ID. Retain it only for an explicitly reviewed retry of that same event; the adapter does not generate IDs or provide durable exactly-once delivery. Identity/message IDs must remain strings, including whole-expression mappings. First-four actions require a user ID or anonymous ID; group requires user/group IDs, and alias requires distinct current/previous IDs. Traits, properties and context use finite named fields. revenueDecimal and valueDecimal accept exact decimal strings (up to 15 integer and six fractional digits); numeric whole-expression values are rejected so a rounded number cannot masquerade as exact text.

configured allows the source's configured destinations; none suppresses destination delivery but still submits data to Segment. Identity/profile/event/group/alias changes can affect audiences, downstream automation, privacy and usage. HTTP 200 records a transport receipt only: output explicitly keeps ingestion and delivery unverified and credentialVerified:false. It does not certify authentication, source ownership or final processing. Timeouts can occur after acceptance; inspect Segment before retrying. There is one request and no automatic retry, batch, arbitrary payload, source management or trigger lifecycle. WORKFLOW_LIVE_SEGMENT and workspace policies govern execution. Provider HTTP is simulated locally; live key validity, ingestion, destination delivery, billing, hosted Review, new-node organization and durable behavior remain untested.

NASA NeoWs workflow nodes

NASA provides three reads: a NeoWs date feed, lookup by exact SPK ID, and the first browse page. Save nasaApi with an existing 40-character apiKey. The fixed api.nasa.gov origin receives it as a query parameter, so requests consume the shared allowance and can appear in provider logs. Test connection explicitly reads the first browse page; it proves that read only, not key ownership or scientific accuracy. See the NASA API portal.

Feed dates must be real calendar dates with an explicit inclusive span of at most seven days. SPK IDs must be exact decimal text, including values mapped from another node; numeric expression values are rejected. Lists return one page with a local disclosure limit of 1–25 (default 10), without automatic paging. Output preserves identifiers/counts as text and returns a finite object summary, rather than scientific datasets, downloads or raw responses. APOD, retired Earth APIs and DONKI are outside this profile. Response projection uses historical NeoWs fixtures; current response compatibility and nullability remain unverified. One request has a 15-second budget and a 2 MiB response cap. WORKFLOW_LIVE_NASA and workspace policies govern execution. Provider HTTP is simulated locally; live key/quota/response behavior, hosted Review, new-node organization and durable storage remain untested.

Metabase workflow nodes

Metabase provides thirteen metadata reads and one explicitly acknowledged saved-question execution. Save metabaseApi with a public HTTPS instanceBaseUrl, optionally including a plain subfolder, and an existing group apiKey beginning mb_. Omit /api, other ports, embedded credentials, query strings and fragments. This profile uses the saved key in X-API-Key; it does not log in, create keys or renew sessions. Test connection reads the saved installation's current API-key user; it does not identify a named human or establish query permissions. See the Metabase API documentation.

Read actions cover databases and their fields, tables, individual fields, questions, metrics, alerts and collections. Returned summaries omit SQL, connection secrets, fingerprints and alert recipients. Use exact decimal text IDs, including expressions. Lists return one bounded page. The collection root value is an output sentinel, not a numeric identifier for arbitrary operations. This finite profile was checked against v0.64.1 source and v0.64 documentation; the unversioned API, deprecated alerts and deployment editions require compatibility testing.

Run saved question sends an empty body to the selected existing question. Explicitly acknowledge execution effects before enabling it. Select output column indexes and a local row limit; duplicate column names remain distinguishable by index. These choices limit disclosed output, not server work: the saved query may consume compute or change datasource data. Output preserves supported scalar types and exact numeric text. No raw SQL, query parameters, export, creation, automatic retry or pagination is available. A lost response can leave the execution outcome unknown; cancellation does not prove rollback. Requests have a 20-second budget, 2 MiB response cap and bounded projected output. WORKFLOW_LIVE_METABASE and workspace policies govern execution. Provider HTTP is simulated locally; live API compatibility, datasource effects, hosted Review, new-node organization and durable behavior remain untested.

Google Tasks workflow nodes

Google Tasks provides twelve actions: list/get/create/change/delete task lists, and list/get/create/change/delete/move tasks or hide all completed tasks. Save googleTasksOAuth2Api with an existing workspace-owned accessToken and explicit declaredAccess of read or write. The declaration is a local restriction, not evidence of Google's grant. Reads require tasks.readonly or tasks at Google; writes require tasks. This manual profile does not offer sign-in, token refresh, service accounts or personal agent credentials. See the Google Tasks REST API.

Choose Save account to save without contacting Google. The saved account remains unverified. Test connection is a separate explicit read of at most one task list; it proves neither identity nor write permission. Replace expired tokens using an authorized workspace account. Blocks cannot change the fixed tasks.googleapis.com destination. Opaque IDs and cursors are encoded once, without following URLs; use exact strings. Only operations documenting the default list accept @default. Lists return one page without automatic paging.

All eight writes require an explicit effects acknowledgement. Deleting a task or an entire list also requires acknowledging possible deletion of assigned originals in Docs or Chat. Hide all completed tasks affects every completed task in the list and requires its own acknowledgement; it does not delete them. Moving a task with omitted parent/previous places it first at the top level. Google enforces assigned and recurring task restrictions. For title and notes fields accepting empty text, edit then clear the control to save an explicit empty string; an untouched required title stays missing. PATCH changes only supplied fields; supported explicit nulls clear notes, due date or completion, and no automatic status/time coupling is added. Due dates retain a calendar date, without a due-time guarantee. A 404 is a failed request, not proof of a previous deletion. Requests have a 15-second budget and 2 MiB response cap, with one attempt and no concurrency or delivery guarantee. WORKFLOW_LIVE_GOOGLE_TASKS and workspace policies govern execution. Provider HTTP is simulated locally; live grants/effects, hosted Review, new-node organization and durable behavior remain untested.

Ordinary n8n v1 task nodes also have bounded import profiles. Source credentials remain inactive until a saved account is explicitly selected. Unknown provider business fields are retained instead of silently discarded; numeric identifiers remain strings and unsafe numeric values fail validation.

Hacker News workflow nodes

Hacker News provides three public reads from the official API v0: get one item, get one user, or read IDs from the top/new/best/ask/show/job feed. No saved account is required. Workspace execution authority and transport controls still apply. This profile uses the official Firebase API; it does not reproduce n8n's Algolia search or import its parameters automatically.

Enter an exact positive decimal item ID as text, or a case-sensitive username. Whole-expression item IDs must resolve to strings; feed IDs can feed a later item node explicitly. Feed offset/limit selects a local slice of one mutable array, without hydration or automatic follow-up requests. The slice is not a stable cursor or historical snapshot. Item results distinguish available, dead, deleted and unavailable; a null response does not establish that an item never existed. Deleted items omit content. HTML and returned URLs remain inert text; no URL is fetched. Long text and reference arrays carry truncation information.

Each action sends one GET to the fixed official host, with one attempt, a five-second budget and a 1 MiB response cap. WORKFLOW_LIVE_HACKERNEWS supplies the deployment default; an explicit workspace transport preference overrides that default. A node retaining an explicit credential reference must still resolve that configuration; the public profile does not silently discard it. Local HTTP is simulated; real API compatibility, hosted Review, organization scope and durable persistence remain unverified.

CircleCI workflow nodes

CircleCI provides ten finite Cloud API v2 actions: project lookup, pipeline list/get/by-number, pipeline workflows, workflow lookup, first-page workflow jobs, legacy OAuth pipeline trigger, definition-based pipeline trigger and workflow cancellation. Save circleCiApi with an existing apiKey and explicit tokenKind:personal. These are personal provider tokens saved as workspace connections; project tokens and automatic token issuance/renewal are unsupported. The declaration does not independently attest the provider token type.

Choose Save account without contacting CircleCI. Test connection explicitly reads /api/v2/me once; it does not establish project access or execution permission. Nodes use the selected saved token on fixed circleci.com, without alternate origins or returned URL downloads. Project slug components, UUIDs and pipeline numbers use their named fields. Whole-expression UUIDs must resolve as strings and pipeline numbers as safe positive integers. List actions return one page; jobs expose the documented first-page profile only.

All three writes require an explicit remote execution acknowledgement and the workflow confirmation path. Triggering runs repository code and may consume credits; cancellation interrupts a workflow and cannot undo earlier effects. The legacy trigger requires an explicit supported OAuth integration and exactly one branch or tag. The definition trigger requires a definition UUID, config/checkout refs and an explicit integration declaration. Parameters are typed name/value rows; false and zero are retained. HTTP 201 records pipeline creation, HTTP 200 from a definition trigger can mean no pipeline was created, and HTTP 202 records a cancellation request rather than a stopped workflow. No automatic retry or polling follows an uncertain outcome, including when the node's Retry On Fail setting is enabled. Check the outcome in CircleCI before deliberately running the write again.

Requests have a 20-second budget, a 1 MiB response cap and bounded projected output. WORKFLOW_LIVE_CIRCLECI supplies the deployment default; an explicit workspace transport preference overrides it. This profile adds no config, artifact or log downloads, triggers or automatic n8n conversion. Local provider responses are simulated; real provider compatibility/effects, hosted Review, organization scope and durable behavior remain unverified.

Brandfetch workflow nodes

Brandfetch provides six finite Brand API projections: metadata, logos, colors, fonts, company and industries. Save brandfetchApi with an existing apiKey. Use a Brand API key rather than a Search client ID. Save account makes no provider request and remains unverified. Test connection explicitly reads Viewer once; it does not establish brand access, endpoint entitlement or available credits.

Choose domain or Brand ID explicitly. Domains use a local ASCII hostname policy and Brand IDs a local id-prefixed policy; URLs, email addresses, arbitrary identifiers and alternate hosts are unsupported. Expressions must resolve to strings. Each action sends one request to fixed api.brandfetch.io with cachedOnly=true. A cached miss returns not-indexed without starting a crawl; an ordinary 404 returns not-found and may report the provider's collection-queued marker without any follow-up. Successful cached reads and ordinary not-found requests can consume credits or paid overage. Cached-only does not establish data freshness, storage rights or asset reuse permission.

Fields distinguish missing from present values, including documented nulls and empty arrays. Output is a finite projection rather than the full provider envelope. Asset URLs remain inert text: nodes do not download or preview logos, fonts, images, HTML or SVG, and do not follow returned URLs. Industries project the company industry list. Search, client-ID browser search, palette generation, batch requests and automatic collection are outside this profile.

Requests use one attempt, a ten-second execution budget and a 1 MiB JSON response cap; cancellation does not prove that no provider work or quota use occurred. WORKFLOW_LIVE_BRANDFETCH supplies the deployment default; an explicit workspace transport preference overrides it. Automatic n8n parameter/credential conversion is unsupported. Local responses are simulated; real provider compatibility, billing, hosted Review, organization scope and durable behavior remain unverified.

Action Network workflow nodes

Action Network provides 39 finite API v2 operations: 26 get/list reads and thirteen writes for people, events, petitions, attendances, signatures, tags and taggings. Other readable resources include forms, surveys, event campaigns, campaigns, advocacy campaigns and fundraising pages. Save actionNetworkApi with an existing apiKey, required declaredListType of unknown, personal or group, and an optional listLabel. The declaration does not verify ownership or permissions. Operations requiring a group list reject other declarations. The key's provider-side list and partner access still apply.

Save account makes no provider request and remains unverified. There is no harmless supported identity check, so Test connection is unavailable and bulk checks skip this account. Nodes use fixed actionnetwork.org/api/v2 routes. Lists return one explicit page of at most 25 records; no returned URL, automatic pagination or hidden resource hydration is followed. UUID inputs must resolve as strings. Contact details and action comments require explicit inclusion and a personal-data acknowledgement; output omits unsupported embedded data.

Each write requires its applicable effect acknowledgements and the workflow's separate confirmation callback. A missing or unsupported confirmation fails before HTTP. Acknowledgements start unchecked; an expression resolving to true is accepted but does not itself attest human consent. Writes may subscribe or resubscribe matching people, propagate subscriptions, replace deduplicated action records, change segmentation or start configured automations. API-created events and petitions are imported records rather than ordinary published registration or signing pages. Review each operation's effects before confirming.

A write acknowledgement does not prove downstream completion. Contradictory response errors or partial-result flags stop the action even when HTTP reports success; informational messages alone do not. These checks are local defensive policy, not a provider-defined error vocabulary. There is no durable single-use approval, expiry or atomic binding to later provider state. Requests have a 15-second provider-phase budget, a 1 MiB response cap, at most 4,096 delivered response fragments including empty ones, bounded local queueing and 300-ms start spacing. These local limits do not coordinate traffic across processes. Actions use one attempt, including when Retry On Fail is enabled; inspect the provider before deliberately repeating an uncertain write. WORKFLOW_LIVE_ACTIONNETWORK supplies the deployment default, with explicit workspace preferences taking precedence. Automatic n8n parameter conversion, key issuance and live triggers are excluded. Provider compatibility, subscriptions, automation effects, hosted Review, organization scope and durable persistence remain unverified; local provider responses are simulated.

Affinity workflow nodes

Affinity CRM provides thirteen finite API v2 operations pinned to 2026-09-17: eleven reads for identity, persons, companies, lists, list entries and field metadata/values, plus two writes for scalar text or number list fields. Save affinityApiV2 with an existing apiKey, positive integer expectedTenantId and exact lowercase expectedTenantSubdomain. This Bearer profile is separate from historical Basic-auth affinityApi credentials; no automatic migration or OAuth renewal is provided. Companies are CRM entities, not the tenant.

Save account makes no provider request and stays unverified. Test connection separately reads whoami once and checks the pinned response version, expected tenant and API grant. It reports status without returning contact details and does not prove write permission. Every nonidentity action checks whoami before its resource request. Calls stay on fixed api.affinity.co; saved subdomains and returned links never select a host. Lists return one page. Entry pages distinguish an explicit null result from an empty array and expose unknown continuation where appropriate.

For a scalar write, read the field first and supply its exact typed expectedValue, or set expectNull and omit that value. Supply the new value, or set clearValue and omit it to send explicit null. Empty text and zero remain values. Required effect acknowledgements start unchecked. After separate workflow confirmation, the action checks identity and reads the exact field again; a changed value, wrong type or nonwritable field stops the POST. Another writer can still change it between GET and POST: this is not atomic compare-and-set, a lock or a durable single-use approval. HTTP 204 records acceptance without a follow-up verification read.

Writes can affect shared CRM data and untested downstream automations. The provider phase has a 20-second budget, each request at most ten seconds within it, a 64 KiB identity cap, 2 MiB resource cap and at most 4,096 delivered response fragments per response, including empty ones. Reads use at most two requests and writes at most three. No automatic retry occurs, including with Retry On Fail enabled; investigate uncertain effects before rerunning. WORKFLOW_LIVE_AFFINITY supplies the deployment default, with explicit workspace preferences taking precedence. Rich field writes, v1 CRUD, bulk operations, automatic pagination and triggers are excluded. Local responses are simulated; live provider grants/effects, hosted Review, organization scope and durable persistence remain unverified.

Adalo workflow nodes

Adalo provides four finite Collections API actions: get a record, list one page, create a record and delete a record. Save adaloApi with an existing Adalo API key and the exact Adalo app ID. Saving makes no provider request and leaves the account saved · unverified. There is no connection test: the account form does not select a collection or authorize a data read.

In each block, select that saved account and supply the matching expectedAppId, collectionId, a positive profileVersion, and one to sixteen declared scalar fields. Each declaration supplies a unique slot, exact propertyName, type of text/number/boolean, and explicit writable flag. Only declared fields are projected; missing, null and scalar values remain distinct. Use reviewed scalar collections: collection classification, complete schema, default values, permissions and downstream effects are not verified. User authentication, relationships and arbitrary JSON values are outside this profile.

Creation uses separate textValues, numberValues and booleanValues slot assignments. Supply every declared writable field and at least one writable field; empty text, zero and false are values, while null assignments are unsupported. Record IDs are exact positive decimal strings within the JavaScript safe-integer range. List defaults to 25 records and accepts at most 100. Its optional continuation object must match the app, collection, selected credential, profile version and page size; a returned continuation is only a candidate, hasMore remains unknown, and the node does not fetch another page automatically.

All four actions require an explicit personal-data acknowledgement. Create and delete also require effect acknowledgements and separate workflow confirmation; creation acknowledges unverified defaults as well. Unchecked controls never authorize a write. These callback checks do not provide durable expiry, single-use approval or atomic credential/request binding. Create reports accepted fields without proving persistence of every field; delete reports HTTP acceptance without proving downstream or cascading completion. A nonempty deletion response must be a valid JSON object without explicit error or partial-result markers; empty success responses remain supported. These are local acknowledgement checks, not a guarantee of the provider’s deletion result. Ordinary declared record fields named error or partial remain data.

Requests stay on https://api.adalo.com, use one attempt even when Retry On Fail is enabled, and do not follow redirects or returned URLs. An uncertain write must be investigated before manually rerunning. Local policies bound each request to 15 seconds, 512 KiB of decoded response, 8192 delivered fragments including empty fragments, and a 32 KiB create body. The process-local four-starts-per-second limiter is not a cross-process quota guarantee. Update, bulk operations, filters, collection management, authentication and token refresh are unsupported. Synthetic local tests exercise the request and response boundaries; live Adalo, hosted Review, organization scope and durable persistence remain untested.

APITemplate.io workflow nodes

Save apiTemplateIoApiV2 with an existing apiKey, then select that saved account in the block. This current v2 profile does not accept historical apiTemplateIoApi credentials automatically. Saving sends no request; Test connection is unavailable. The explicit account-information node reads bounded quota values and does not verify identity, ownership or generation permissions.

Five operations read account information, list one template page, list one generated-object reference page, create a PDF or create an image. PDF inputs use up to sixteen typed slots in the JSON editor: each has name, type (string, number, boolean or null) and value. Image inputs use up to sixteen name/text override rows. Empty text, zero, false, null and omission remain distinct; this finite profile does not accept arbitrary template JSON, HTML or URL rendering. Lists return identifiers rather than names, content or download URLs and never paginate automatically.

Generation requires explicit acknowledgements for quota, external assets and CDN storage plus separate workflow confirmation. Requests ask for 60-minute expiry, CDN storage and disabled post-actions; these settings do not prove actual retention or downstream effects. Requests stay on https://rest.apitemplate.io/v2, never follow redirects or returned URLs and make one attempt even with Retry On Fail enabled. Local execution limits are nine seconds after credential/confirmation preparation, 64 KiB and 2048 response fragments. The current process-local policy remembers at most 32 saved account references. Any dispatched failure, including a failed read, retires that reference in that process; there is no reset control, and using it again requires the worker module to reload. These are conservative local limits rather than provider requirements. A cancelled generation may already have applied, even when the error only says it was cancelled. Investigate generated references in the provider before considering another invocation; a retired account cannot be rerun in the same process. Live generation, retention, hosted Review, organization scope and durable persistence remain untested.

Airtop workflow nodes

Save airtopApiKey with an existing apiKey. This native profile does not accept historical airtopApi credentials automatically. Saving sends no request and Test connection is unavailable. Choose an existing sessionId explicitly; get-session, list-windows and close-window do not discover or provision sessions. Closing also requires the exact windowId.

Every operation requires acknowledgeSensitiveMetadata and separate workflow confirmation. Closing additionally requires acknowledgeBrowserControl: it can discard unsaved browser state and does not terminate the session or stop billing. Outputs contain only bounded session status/usage or window IDs; titles, page URLs, CDP addresses and raw configuration are omitted. A close acknowledgement is not independent proof that the browser was destroyed.

Requests use https://api.airtop.ai/api/v1, one attempt, no redirects and no URL following. Local provider execution is bounded to fifteen seconds after credential/confirmation preparation, 64 KiB, 128 fragments and fifty window IDs. One active request and one start per second per selected account are process-local policies. Provisioning, termination, navigation, live view, AI, scripts, extraction, files and CDP are unsupported. A dispatched close may already have applied after cancellation; investigate before rerunning. Real browser control, billing, hosted Review, organization scope and durable persistence remain untested.

Autopilot Journeys workflow nodes

Save autopilotApi with an existing apiKey for the legacy Journeys API at https://api2.autopilothq.com/v1. This is separate from Ortto CDP. Current availability of the legacy service and its credentials remains unverified. Saving sends no request and Test connection is unavailable.

Eight operations get or upsert a contact, list lists or published triggers, check contact/list membership, add or remove list membership, and add a contact to a published journey trigger. Reads require acknowledgeRead. The four writes also require acknowledgeEffects and separate workflow confirmation because they can change subscriptions, run journeys or send configured messages. Names and titles are omitted unless both includeDetails and acknowledgeDetails are explicitly enabled. A restricted membership-not-found response does not distinguish a missing contact from a missing membership; it is not proof that the contact exists.

Use the exact typed IDs and explicit supported fields; supplied nulls and omitted values keep their operation-specific meanings. To clear a supported standard field with null, map a typed expression such as {{input.clear}} whose value is null. Typing the word null in a text field sends text; the parameter inspector does not round-trip a manually authored top-level null as a distinct control value. Typed custom-field JSON keeps its explicit nulls. No webhook trigger, arbitrary contact fields, bookmark pagination, OAuth renewal or automatic n8n conversion is added. Local limits are one request per invocation, five seconds of provider execution after credential/confirmation preparation, an 8 KiB request, a 64 KiB response, 256 fragments and fifty collection entries. There are no automatic retries or redirects. A cancelled write may already have applied despite a generic cancellation error; check the exact contact, list or journey before another invocation. Live legacy behavior, messaging effects, hosted Review, organization scope and durable persistence remain untested.

Acuity Scheduling workflow nodes

Save acuitySchedulingApi with the existing Basic authentication userId and apiKey, then select that saved account in each block. User ID is exact positive decimal text; the maximum 2147483647 is this finite integration's local policy, rather than a documented vendor maximum. Keys use 1–512 visible ASCII characters without whitespace. Saving sends no request and Test connection is unavailable. The explicit account-get action reads bounded metadata and checks the authenticated User ID against the selected saved account; it does not certify appointment permissions.

Seven actions read account, calendar or appointment-type metadata, list available dates or times, cancel an exact appointment, or reschedule one. Every action requires acknowledgePersonalData and separate workflow confirmation. Cancellation and rescheduling also require acknowledgeEffects, acknowledgeNotifications, acknowledgeCosts, the exact integer appointmentID, and an explicit boolean noEmail. IDs remain typed integers where required; dates use the declared calendar format and times include an offset. An availability result is a snapshot, not a reservation. Client rules apply; no administrative bypass is sent. Cancellation cannot be undone through this endpoint. noEmail=true requests suppression of that action's email/SMS only; reminders, downstream effects, fees and refunds remain unverified.

Outputs omit client/staff names, emails, notes, forms and URLs. The account User ID is an identifier and does not redact matching digits from dates or prices. As a conservative local policy, a projected string containing an actual API-key or Basic-header secret stops the entire result instead of returning altered typed data; a dispatched write remains uncertain. Requests stay on https://acuityscheduling.com/api/v1, use one attempt even with Retry On Fail enabled, and never paginate or follow returned URLs automatically. Local limits are 64 KiB, 128 response fragments, 100 collection items, one active request and one start per second per selected account. Provider execution has a ten-second deadline after credential and confirmation preparation. A cancelled or uncertain write may already have applied; inspect the exact appointment before considering another invocation. The native REST actions add no Acuity webhook trigger, OAuth flow or automatic n8n import conversion. Live provider behavior, hosted Review, organization scope and durable persistence remain untested.

Bannerbear workflow nodes

Save bannerbearApiV5 with an existing apiKey for the current V5 API. V5 and V2 keys are incompatible; the native profile has a separate credential type. Saving sends no request and Test connection is unavailable. Select that saved account explicitly in each block. Reads require acknowledgePii and separate workflow confirmation because template names can contain personal information.

Four actions create an image, read one image status, read one template, or read one template page. Creation needs an exact templateUid, one to 32 unique objects modification rows and explicit acknowledgeQuota, acknowledgeExternalAssets and acknowledgeStorage, alongside the data acknowledgement and confirmation. Supported scalar modifications include text, color, backgroundColor, hidden and opacity. Empty text, hidden=false, opacity=0 and transparent=false retain their types; untouched optional fields stay omitted. Optional formats use an explicit finite JSON array. The existing template may fetch assets or invoke configured AI; generation can consume quota or money and store supplied text and generated assets. Public visibility, retention and downstream effects remain unverified.

Requests use the asynchronous https://api.bannerbear.com/v5 host only. HTTP 202 acknowledges acceptance; image status remains pending, completed or failed. Read status explicitly in a later node; no automatic polling, sync fallback, file download, URL following or webhook setup is added. Outputs contain finite IDs/status or template metadata, with raw design text, provider diagnostics and asset URLs omitted. Local limits include a 256 KiB response, 2048 fragments, at most 25 templates per explicit page, one active request and twelve starts per minute per selected account. Provider execution has a ten-second deadline after credential/confirmation preparation. Each invocation makes one attempt even with Retry On Fail enabled; a lost or cancelled create may already have been accepted. Inspect the provider before considering another invocation. Local replies are simulated; live rendering, billing, hosted Review, organization scope, durable persistence and n8n parameter conversion remain untested.

Beeminder workflow nodes

Save beeminderApi with an existing personal authToken, then explicitly select that account in the block. The native profile does not accept OAuth credentials. JSON must contain exactly one nonempty authToken string with valid Unicode, no ASCII control characters and at most 2048 UTF-8 bytes; these are local admission limits. Saving sends no request, leaves authentication unverified and offers no connection test. Requests use https://www.beeminder.com/api/v1/users/me only. Beeminder authentication puts the token in a GET query or a URL-encoded POST body; keep request URLs and bodies out of shared logs.

Three actions read finite goal metadata, read one datapoint page, or create one datapoint for an exact goalSlug. All require acknowledgePersonalData and separate workflow confirmation. Creating also requires acknowledgeEffects and acknowledgeFinancialConsequences: changed progress can affect commitments, derailment and pledge billing even though the node calls no charge endpoint. Enter value as a number, including zero or negative values; optional timestamp uses integer Unix seconds. Editing and clearing optional comment preserves an empty string; leaving it untouched omits it. Returned values describe the provider reply and do not prove that requested value/time was echoed unchanged.

List page and per remain typed integers. One invocation reads one response with at most 25 datapoints; there is no automatic pagination. The provider's per=0 option requests all datapoints, but this finite profile rejects a reply exceeding its local 25-item cap. Output omits goal names, URLs and raw diagnostics; comments can contain personal data. Missing optional fields remain absent, declared nullable goal scalars preserve null, and false/zero/empty values retain their types. A projected string overlapping a raw or declared encoded token form rejects the entire result as a conservative local policy.

Local limits include a 256 KiB response, 2048 fragments, depth 12, 8192 JSON nodes, absolute numeric values up to 1e12, timestamps from 0 through 4102444800, one active request and 250 ms between dispatches per selected saved account. Provider execution has a ten-second deadline after credential/confirmation preparation. Each invocation makes one attempt even with Retry On Fail enabled; failed reads allow a corrected deliberate later invocation. A lost or cancelled create may have applied, including when the gateway returns a generic abort message. Inspect the exact goal before another deliberate creation. No requestid upsert, goal editing/deletion, pledge/charge endpoint, automatic retry, redirect, polling, OAuth renewal or n8n conversion is added. Real-provider effects, hosted Review, organization scope and durable persistence remain untested.

Chargebee workflow nodes

Save chargebeeApiV2 with an existing site and apiKey. Site is one lowercase label, without a URL or .chargebee.com suffix. Choose the intended test or live site's key explicitly. Advanced JSON accepts exactly {"site":"your-site","apiKey":"YOUR_KEY"}; the historical n8n chargebeeApi/accountName format is unsupported. Saving sends no request and Test connection is unavailable. Select the saved account in each block; requests stay on that site's https://<site>.chargebee.com/api/v2 origin and use Basic authentication with the API key as username and an empty password.

Eight actions read customer metadata, create a customer, list one invoice page, read an invoice, obtain a sensitive invoice PDF link, read a subscription, cancel it with explicit current Product Catalog 2.0 choices, or schedule its deletion. All require acknowledgePersonalData and separate workflow confirmation. Creation also requires acknowledgeCreate, an explicit customer id and autoCollection on/off choice; configured notifications and connected CRM records may follow. Invoice lists use a typed limit from 1–100, optional includeDeleted and an opaque offset. Omission uses the local defaults 10 and false; reuse nextOffset explicitly for another page. Outputs omit contact, address, card, payment and custom metadata. A consolidated invoice's subscription ID may be null.

The PDF action additionally requires acknowledgeCapability. Its HTTPS downloadUrl is a sensitive capability that this workflow can expose; the action never downloads it. Use returned validTill when present and do not assume a fixed lifetime. Cancellation requires acknowledgeCancellation, an explicit mode (no-contract/contract) and the matching cancelOption or contractTermCancelOption. A fresh subscription read checks the selected ID, status and contract presence before at most one cancellation POST; concurrent changes remain possible. Specific-date cancellation needs a future Unix timestamp cancelAt, checked again after that read. Immediate cancellation requires all four explicit financial choices: creditOption, unbilledChargesOption, accountReceivablesHandling and refundableCreditsHandling. Scheduled modes reject those fields. Site policies, default contract termination fees and deletion of future ramps can apply; unsupported scheduling modes can be rejected by the site.

Deletion requires acknowledgeDelete and acknowledgeDeletionPrerequisites after external review of consolidated invoices and cross-subscription credits. It schedules permanent deletion and can cascade to invoices, credit notes, transactions and usages, affecting reports and integrations; associated usage events remain. deletion_scheduled does not mean deletion has completed. This action does not verify or clean up prerequisites.

Each invocation makes one attempt despite Retry On Fail; cancellation alone uses one GET followed by at most one POST within the same ten-second provider execution budget, after credential and confirmation preparation. Local response limits are 256 KiB, 256 fragments, depth 12 and 8192 JSON value/key nodes. IDs, dates and amounts use the declared finite types; omit optional fields rather than supplying null or undefined. Known authentication-secret string collisions stop the whole result before redaction. No redirects, automatic pagination, polling, reconciliation or n8n import conversion is added. A lost or cancelled POST may have applied, including when the gateway reports a generic abort; inspect the exact resource before rerunning. Real billing, site permissions, deletion prerequisites/completion, hosted Review, organization scope and durable persistence remain untested.

Bubble workflow nodes

Save bubbleDataApi with the exact appSlug, explicit environment (live/development), declared tokenClass (user/admin) and existing apiToken. Advanced JSON requires all four string fields, for example {"appSlug":"your-app","environment":"development","tokenClass":"user","apiToken":"YOUR_TOKEN"}. There are no environment or token-class defaults. Saving sends no request and Test connection is unavailable. The native profile uses https://<appSlug>.bubbleapps.io/api/1.1/obj, with /version-test before /api for development. Custom domains and branches are unsupported; historical bubbleApi credentials are not converted automatically.

Five actions get one object, list one page, create, PATCH selected fields, or delete an object. Select the saved account and set expectedAppSlug and expectedEnvironment to its exact values; these fields cannot override its host or environment. Every action requires acknowledgePersonalData and separate workflow confirmation. Declared admin tokens additionally require acknowledgeAdminPrivacyBypass for every action because they may bypass privacy rules. Token class is a user declaration, not proof of privileges. Writes require acknowledgeEffects and deletion additionally requires acknowledgeDestructive.

Use an explicit typeName and 1–16 declaredFields rows with exact API field names, type (text/number/boolean) and explicit nullableRead. User/account types and reserved credential/system fields are unsupported. Reads also need an explicit selection from those declarations. Missing fields remain omitted; null is returned only for a declared nullable read. Writes use rawFieldsJSON, a strict JSON object containing 1–16 declared scalar fields. Enter literal JSON; expressions in rawFieldsJSON are rejected by the current workflow readiness check. Empty text, zero and false remain values; null writes and duplicate JSON keys are rejected. Do not place credentials in custom fields. Declared fields can still contain personal data, and a create ID does not prove every supplied field was persisted.

Lists require typed cursor and limit (1–100), permit up to eight typed filters, and accept an optional sort with explicit boolean descending. The local cursor range is 0–49999 and cursor+limit cannot exceed 50000; a continuation beyond that boundary is rejected. Reuse nextCursor deliberately for another page. No automatic pagination, returned-URL following or additional read after a write occurs. Custom relationships, files, dates, lists and arbitrary JSON fields are unsupported.

Each action makes one attempt despite Retry On Fail. Local limits include a 64 KiB serialized request, a 1 MiB response, 2048 fragments, depth 8, 8192 JSON nodes and ten seconds of provider execution after credential/confirmation preparation. A known authentication-secret collision in selected strings or field names stops the whole result before gateway redaction; this does not recognize every possible encoding. A dispatched write can remain uncertain after failure or cancellation, including a generic gateway abort. Inspect the object before another deliberate write; defaults, downstream app effects, privacy rules and recovery are not guaranteed. This profile adds no bulk action, custom workflow endpoint, OAuth issuance, live trigger or automatic n8n conversion. Live provider behavior, hosted Review, organization scope and durable persistence remain untested.

Currents.dev workflow nodes

Save currentsApi with an existing organization API key for Currents.dev test analytics. This integration does not use the Currents news API. Advanced JSON accepts exactly {"apiKey":"YOUR_ORGANIZATION_KEY"}. The local key format is 1–2048 visible ASCII characters without whitespace; personal access tokens beginning crnts_pat_, OAuth credentials and generic credential fallbacks are unsupported. Saving sends no request, leaves authentication unverified and offers no connection test. Select the saved account in each block; requests use Bearer authentication on https://api.currents.dev/v1 only. Key format does not verify organization membership or permissions.

Three actions read one project page, one exact projectId, or one project's run page. Every action requires acknowledgePersonalData and separate workflow confirmation because names, searches and test metadata can contain private organization data and reads can consume quota. Project IDs are exact strings encoded once as a path segment. Lists accept typed limit from 1–25, defaulting to 10, and at most one of startingAfter or endingBefore. Leaving a cursor untouched omits it; editing and clearing it sends an explicit empty cursor. Run lists also accept search, status and completionState. Search must contain non-whitespace printable ASCII text, and its surrounding spaces are transmitted unchanged.

Outputs retain finite project IDs/names/dates and declared run scalars. Raw commits, author metadata, attachment URLs and provider diagnostics are omitted. Run fields are optional: an empty object or has_more: true without a returned cursor does not invent a run ID or continuation. Optional nulls, false, zero, empty text and omitted fields remain distinct where declared. Reuse a returned cursor deliberately in another invocation; there is no automatic pagination. A selected output string or field name overlapping the complete known raw or declared encoded API key fails the whole result before redaction.

Local bounds include a 256 KiB exposed response, 2048 fragments, depth 12, 8192 JSON nodes and the requested page limit. Each invocation makes one attempt despite Retry On Fail. Provider execution has a ten-second budget after confirmation and credential preparation; that preparation has no native deadline. Failed reads can be invoked deliberately later, with one active request and 250 ms between requests per saved account. No trigger, mutation, PAT issuance, OAuth renewal, polling or automatic n8n conversion is added. Real provider access, hosted Review, organization scope and durable persistence remain untested.

CoinGecko workflow nodes

Save coinGeckoDemoApi with an existing Demo API key. Advanced JSON accepts exactly {"apiKey":"YOUR_DEMO_KEY"}; the local format is 1–2048 visible ASCII characters without whitespace. Pro keys, historical credential aliases, OAuth, keyless fallback and custom hosts are unsupported. Saving sends no request, leaves authentication and plan access unverified, and offers no connection test. Select the saved account explicitly. Requests use https://api.coingecko.com/api/v3 with the x-cg-demo-api-key header. Every invocation requires separate workflow confirmation because authenticated reads can consume quota; this public-market profile adds no second acknowledgement checkbox.

Three actions list supported quote currencies, read USD prices for explicit coin IDs, or read one USD market page. The currencies action has no input fields. Prices require ids, a JSON array of 1–25 unique exact IDs, for example ["bitcoin","ethereum"]; comma-separated text is rejected. IDs use the local lowercase letter/digit/hyphen grammar and at most 96 characters. Markets require typed integer page from 1–1000 and perPage from 1–25, without implicit defaults. Complete typed expressions can supply the ID array and numeric inputs; a string or invalid resolved value fails before its request. Quote currency is USD, market ordering is descending market capitalization, and there is no automatic pagination.

Price output contains only returned requested IDs and their declared USD scalars. A missing coin or field remains absent, an unknown-only row can remain empty, zero and negative values stay numeric, and the declared 24-hour change can be null. Market output retains finite ID/symbol/name, price/market/rank/volume/change scalars and the supplied timestamp. Declared nullable market scalars preserve null. Image URLs, ROI, sparkline and raw diagnostics are omitted. Responses do not certify completeness, current prices or freshness; inspect the returned timestamp where available. An exact known API-key collision in a selected string or actual object key rejects the entire result before redaction.

Each invocation makes one attempt despite Retry On Fail. Local bounds include a 256 KiB exposed response, 4096 yielded fragments, depth 12, 50000 JSON nodes and a 64 KiB projected output. Provider execution has a ten-second budget after confirmation and credential preparation; the native gateway does not bound that preparation. Invalid UTF-8, decoded duplicate JSON keys and responses outside the finite contract fail as a whole. No trade, mutation, trigger, Pro plan, automatic retry, returned-URL following or n8n parameter conversion is added. Live authentication, vendor quotas, hosted Review, organization scope and durable persistence remain untested.

Disqus workflow nodes

Save disqusServerApi with an existing server API secret. Advanced JSON accepts exactly {"kind":"apiSecret","apiSecret":"YOUR_SERVER_SECRET"}; the structured form supplies the fixed kind and asks for one password field. The local secret format is 1–4096 visible ASCII characters without whitespace. Public keys, the historical n8n Access Token credential, OAuth and generic credential fallbacks are unsupported. Saving sends no request, leaves authentication unverified and offers no connection test. Select the saved native account explicitly. Requests use https://disqus.com/api/3.0 with the secret in the api_secret query parameter; do not share request URLs.

Four actions read one forum, or one page of its threads, posts or categories. Supply the exact forum shortname. Every action requires acknowledgePersonalData and separate workflow confirmation. Lists accept typed limit from 1–100, defaulting to 25, optional order asc/desc and an opaque nonempty cursor. Omitted order uses descending for threads/posts and ascending for categories. Reuse a returned cursor deliberately for another invocation; next and hasNext are independent and never trigger a request. There is no automatic pagination or returned-link following.

Outputs contain finite IDs and declared metadata. Post raw_message remains untrusted text and can contain personal data, email addresses or HTML text. Structured author, email, IP, media and formatted-message fields are omitted; this is not a promise that every returned string is free of personal data or HTML. Declared nulls, false, zero, empty text and missing optional fields retain their meanings. An exact known authentication-secret collision in a selected string or actual object key rejects the entire result before redaction.

Each invocation makes one GET attempt despite Retry On Fail. Local limits include a 1 MiB exposed response, 1024 read results including the final done result, depth 16, 10000 JSON nodes, 1000 entries per container and 262144 aggregate string units. Provider execution has a ten-second budget after preparation; the gateway allows eleven seconds after confirmation and credential preparation, without bounding that preparation. No write, moderation, SSO, OAuth issuance, trigger or automatic n8n conversion is added. Real authentication, forum access, vendor quotas, hosted Review, organization scope and durable persistence remain untested.

DHL workflow nodes

Save dhlApi with an existing production DHL Tracking subscription API key. The form asks for one password field; Advanced JSON accepts exactly {"apiKey":"YOUR_SUBSCRIPTION_KEY"} with 1–2048 visible ASCII characters and no whitespace. Saving sends no request and leaves authentication, permissions and shipment ownership unverified. Test connection is unavailable. Select the saved native DHL account explicitly.

Choose Read shipment tracking and enter the exact trackingNumber. Leave recipientPostalCode omitted unless you intend to request postal enrichment: it can unlock additional private shipment or proof-of-delivery data even though this node omits those detailed fields. Spaces, plus signs and leading zeros are preserved. An explicitly empty postal code is invalid. Acknowledge the personal-data read and confirm the invocation separately.

Optional typed offset accepts 0–1000 and defaults to 0; limit accepts 1–5 and defaults to 5. These are local bounds. Each invocation makes one GET attempt, even with Retry On Fail enabled, with no automatic pagination, polling or returned-link following. Requests use the production Tracking endpoint at https://api-eu.dhl.com/track/shipments and the DHL-API-Key header. Reads consume your subscription quota.

Results contain a bounded shipment list with optional ID, service, return flag, date strings, status codes and event timestamps. Dates and future status/service codes stay raw text without calendar validation or timezone conversion. Returned IDs may differ from your input when the service resolves a piece or reference. Missing fields, empty values and false return flags retain their meanings. Names, addresses, locations, proof of delivery, signatures, links and free-form diagnostics are omitted. Tracking metadata can still contain personal data; an exact saved-key collision in a selected string or object key rejects the entire result.

The node accepts at most five requested shipments and 100 events per shipment in a 256 KiB exposed response. Local spacing is five seconds per saved credential ID; it does not enforce whole-key, daily or cross-process provider quotas. Execution has a ten-second budget after preparation, with an eleven-second gateway budget after confirmation and credential preparation; that preparation is not time-bounded. No write, OAuth, sandbox fallback, trigger or automatic n8n conversion is added. Real provider authentication, shipment access, hosted Review, organization scope and durable persistence remain untested.

Cockpit workflow nodes

Cockpit has two native actions: list a first page of content items, or read one model item. This profile follows the Cockpit 2.14.1 content API. It uses a saved cockpitContentApi connection with your root HTTPS origin and content API key. Legacy query-token credentials and routes are not supported.

Save the origin exactly as a root origin, such as https://cms.example.com, without a trailing slash, path or port. Use a content key with access to the model you intend to read. Saving sends no request and leaves access unverified; Test connection is unavailable for this profile.

Choose the model and 1–8 distinct field names explicitly. List selected content fields also requires a limit of 1–25 and reads the first page only. Get selected content item fields returns one projected item or the provider's literal false; it does not prove that the model is a singleton. Missing selected fields remain absent. Selected text, number, boolean and null values are preserved; nested selected values cause the action to fail. Other content and the total count are omitted.

Review the selected fields before checking the personal-data acknowledgement. The checkbox starts unchecked, and each execution also requires a separate confirmation. Selected content can still contain private information. Each action sends one request, with no automatic pagination, redirect following or retry. Population, arbitrary filters, writes and anonymous access are outside this profile.

The exposed decoded response is limited to 256 KiB, the selected output to 64 KiB, and reading to 4096 iterations including completion. Provider execution has ten seconds after preparation, with an eleven-second gateway budget after confirmation and credential preparation; that preparation has no deadline. These are local support limits. Real Cockpit authentication, model privileges, hosted Review, organization scope and durable persistence remain untested.

Drift workflow nodes

Use a saved driftApi account containing an existing private-app access token. The structured form has one required Drift private-app access token field. Save stores the credential locally without contacting Drift; authentication, granted scopes, identity and expiry remain unverified. Test connection is unavailable. Generic HTTP credentials, OAuth connections and names that merely start with DRIFT_ cannot substitute for this native account.

Two reads are available:

• Read contact by ID requires the contact ID as an exact decimal string and an explicit JSON array of attribute names, for example ["email","company"]. Use [] to return only the ID. IDs support the full signed 64-bit range, from "-9223372036854775808" through "9223372036854775807"; numeric workflow inputs, exponent notation, fractions and leading zeros are rejected. The response ID must match the requested string exactly.
• List contact attribute definitions reads bounded name, displayName and type metadata. This requires the elevated all_contact_read grant. The token's text does not establish that it has this grant. Returned type labels remain raw metadata; they do not convert contact values.

Select the saved account, enable Drift live reads in workspace preferences, acknowledge access to personal or internal organization data, and confirm each invocation separately. The acknowledgement starts unchecked. Selecting an account, enabling transport or saving parameters does not provide that confirmation.

Contact reads return only selected scalar attributes. Missing fields remain absent and appear in missingAttributeNames; present null, false, zero and empty strings retain their values. Selected objects, arrays, nonfinite numbers or integer attribute values outside the safe JavaScript integer range fail the read. The exact 64-bit precision guarantee applies to the contact ID. Unknown or unselected contact fields are omitted.

Each invocation makes one request and one attempt, including when Retry On Fail is enabled. Responses are limited to 1 MiB, selected attribute names to 32 and definitions to 200. Selected strings are limited to 4096 code units. These are local support bounds; larger responses fail without partial output. Provider execution has ten seconds after preparation, with an eleven-second gateway budget after confirmation and credential preparation; preparation has no deadline.

Writes, email/external-ID lookup, contact lists, automatic pagination, OAuth refresh and n8n import aliases are unsupported. The current verification uses local synthetic responses. Real Drift authentication and token grants, hosted Review, organization scope and durable persistence have not been tested.

E-goi workflow nodes

Use an existing API key saved as egoiApi. The native credential form has one password field, E-goi API key. Saving sends no request to E-goi; authentication, account identity, permissions and expiry remain unverified. Test connection is unavailable. Select the saved native account and enable E-goi live reads. Generic HTTP credentials, OAuth and credentials saved with the generic type egoi cannot authorize these nodes.

Both operations start with the personal-data acknowledgement unchecked and require a separate confirmation before each invocation. Acknowledgement is never supplied automatically.

List IDs support values through 9223372036854775807; keep them as strings or typed string expressions to preserve precision. The limit and offset accept typed numeric expressions. These are local support bounds. Contact ID letter case is preserved; provider case normalization has not been verified.

Mailing-list results select list_id, internal_name, public_name, status, group_id, created and updated. Contact results select contact_id, status, email, first_name, last_name, language, email_status and subscription_date. Missing optional fields stay absent; null and empty strings remain distinct. Numeric list IDs and optional counts retain their original decimal values as strings, including zero. Other fields are dropped after the bounded response has been read; this is not a provider-side field restriction.

Each invocation sends one GET request to https://api.egoiapp.com with API-key authentication. Retry On Fail does not repeat either read. Redirects, arbitrary URLs, email search, account discovery, writes, sends, enrollment, OAuth refresh, import aliases and automatic pagination are unsupported. A full page does not establish that all records were returned.

The response is bounded to 256 KiB of decoded data, 1,024 body reads including completion, and a 10-second provider deadline after confirmation and credential preparation. The public result is bounded to 64 KiB. Invalid, oversized or secret-echoing responses fail with a fixed local error instead of returning partial data.

The strict credential parser checks the text it receives before parsing, up to 32,768 JavaScript string units. Settings trims the outer text first, so this does not cap the original editor input. Registered Settings validation and direct storage are distinct: malformed nonempty text can be stored unverified through direct persistence, but execution validates it before any provider request.

Local responses and credentials used for verification are synthetic. Real E-goi authentication, permissions, case behavior, live DNS/TLS, Review, organization authorization and durable storage remain untested.

Elastic Security workflow nodes

Save an existing Kibana API key as elasticSecurityApi using the raw JSON editor:

{
  "apiKey": "<existing canonical Base64 API key>",
  "baseOrigin": "https://kibana.your-domain.com",
  "spaceId": null,
  "apiVersion": "v9"
}

Replace the API-key placeholder with the raw Base64 key, without an ApiKey prefix. All four keys are required. Use JSON null for the default space, or an explicit lowercase space ID; the string default is not accepted. The origin must be a lowercase root HTTPS DNS origin without a trailing slash, path or explicit :443. Native :9243 is also supported. Local validation does not establish key privileges, server version or public routability. The existing DNS restrictions and pinned TLS transport still apply.

Saving sends no provider request; Test connection is unavailable. Settings validates the JSON after outer whitespace trimming. Other direct persistence paths can save malformed nonempty text, which execution rejects. Select the saved native account and enable Elastic Security live reads. Generic HTTP credentials, OAuth and a generic elastic-security credential cannot authorize these operations.

Both operations start with privacyAcknowledged unchecked and also require independent gateway confirmation. Typed whole expressions retain their original value types until execution. Unknown operation fields are rejected without printing their names in template or placeholder validation errors.

List reads request owner=securitySolution and return one page. Get checks the returned owner after reading; it cannot guarantee server-side owner filtering. Full private case content can reach the bounded parser before title, description, identities, comments and nested content are discarded. Returned metadata can still be personal or sensitive.

Selected case fields include ID, owner, status, severity, timestamps and bounded counts. Missing optional counts stay absent; permitted nulls and zero remain distinct. Counts require original unsigned decimal integer tokens through 9007199254740991; decimal and exponent spellings are rejected before numeric conversion. Responses are limited to 512 KiB of Fetch-exposed decoded bytes. Retained known compression headers do not trigger another decompression.

Each invocation makes one attempt, including when Retry On Fail is enabled. There are no redirects, automatic pages, writes, triggers, version negotiation, OAuth refresh or n8n import aliases. Local verification uses synthetic responses. Real authentication, privileges, DNS/TLS, deployed versions, organization authorization, durable storage and the Review walkthrough remain untested.

Emelia workflow nodes

Save an existing raw API key as emeliaApi in the Emelia API key password field. The key must be exactly 48 visible ASCII characters, without a Bearer prefix. The field preserves what you type; spaces are rejected rather than trimmed from the key. Saving sends no provider request; Test connection is unavailable. Identity, permissions and expiry remain unverified. Settings validates after outer JSON whitespace trimming; other direct persistence paths can store malformed nonempty text unverified. Execution validates the selected native saved account before dispatch. Generic HTTP credentials, OAuth, name matching and a generic emelia type cannot substitute.

Personal-data acknowledgment starts unchecked, can be unchecked again, and is separate from the confirmation required for each invocation. Review the saved account and exact read before confirming. Whole expressions preserve boolean, number and string types. Names, dates and operational counts may contain personal data. Full responses, including nested campaign, sequence, contact, CRM and integration internals, reach the bounded parser before local scalar projection; this is not server-side field selection. Missing fields stay absent; supported null, false, zero and empty strings remain distinct.

Campaign IDs remain strings with their spelling preserved; get compares the returned ID case-insensitively. List IDs are opaque visible ASCII strings, not numeric or necessarily hexadecimal. Campaign listing has no server pagination: the local 100-record limit rejects a larger result without truncation or an invented server limit. Lists send both page and limit and never follow returned URLs or automatically request another page. Reported pageSize need not equal requested limit or items length.

Selected counts accept original canonical unsigned decimal JSON integers through 9007199254740991, compared before numeric conversion. Fraction, exponent, signed-zero and unsafe counts are rejected locally. Date strings use a bounded uppercase date-time grammar with Gregorian date validation; unknown campaign statuses and wider vendor-valid dates, numbers or collections may be unsupported. Decoded bodies are limited to 1 MiB, 4,096 reads including completion, depth 16 and 20,000 values/containers; projected JSON is limited to 64 KiB. Invalid or oversized responses fail without partial output. Known compression headers do not trigger another decompression or decoded-versus-wire length comparison.

Enable Emelia live execution for the workspace; a saved workspace boolean overrides the runtime deployment default from WORKFLOW_LIVE_EMELIA. Separate deployment catalog policies can still make a block unavailable. Requests use fixed https://api.emelia.io, raw Authorization, one attempt and no redirect following. Retry On Fail does not repeat a read. Provider 10-second and gateway 11-second execution budgets begin after acknowledgment, independent confirmation and credential preparation; they do not bound preparation itself. Saving parameters or credentials and declining confirmation send no request.

These native REST reads do not convert legacy n8n GraphQL nodes or add import aliases, writes, campaign starts, contact insertion, discovery or OAuth refresh. Local tests use synthetic responses. Real Emelia authentication, permissions, provider shapes, DNS/TLS, matching-branch Review, organization authorization and durable storage remain untested.

ERPNext workflow nodes

Create a native erpNextApi connection using the raw JSON editor. Enter raw JSON with exactly apiKey, apiSecret, environment and domain; cloud-hosted connections also require subdomain. Both secrets must contain 1–4096 visible ASCII characters without a colon. Saving sends no request and does not verify identity, roles or permissions; Test connection is unavailable.

For cloud hosting, use environment: "cloudHosted", one lower-case DNS label in subdomain, and domain: "erpnext.com" or "frappe.cloud". For self hosting, use environment: "selfHosted" and a canonical lower-case root HTTPS origin such as https://erp.company.com, without a trailing slash, explicit port, path, query or fragment; omit subdomain. For example:

{
  "apiKey": "<existing key>",
  "apiSecret": "<existing secret>",
  "environment": "cloudHosted",
  "subdomain": "office",
  "domain": "frappe.cloud"
}

Or, for self hosting:

{
  "apiKey": "<existing key>",
  "apiSecret": "<existing secret>",
  "environment": "selfHosted",
  "domain": "https://erp.company.com"
}

Replace the placeholders with existing credentials. These secret character limits are local support bounds, not a vendor key format. Only the optional extra allowUnauthorizedCerts: false is supported. TLS verification remains enabled. Do not use passwords, cookies, Basic credentials or OAuth as a substitute for the saved native API key and secret.

Select that saved native connection on List one metadata page or Get one document metadata projection. Both use the v1 /api/resource/ API. Enter an exact DocType of 1–80 ASCII characters: begin with a letter, then use letters, digits or underscores, with single spaces between words; the reserved DocType discovery resource is unsupported. Listing uses limit (1–100, default 10) and offset (0–1,000,000, default 0). Get requires the exact original document name of 1–140 ASCII characters: begin with a letter or digit, then use letters, digits, spaces, underscore, dot, @, colon, plus or hyphen, without a final space. Case and leading zeros remain intact; Unicode names, slash, percent escapes and query/fragment delimiters are unsupported. No operation loads resource options, discovers fields, follows another page or accepts arbitrary filters, SQL, RPC or fields.

Privacy acknowledgment begins unchecked and can be unchecked again. Review it and separately confirm the selected account and read. A Get downloads the full bounded document, potentially including personal information and child tables, before local projection. Returned metadata contains name and supported optional creation, modified and docstatus fields. Empty text, null, zero and absent fields remain distinct. Status accepts original integer tokens 0, 1 or 2; names remain strings. List limit and offset echo request settings and do not establish totals or completeness.

Requests use the saved canonical HTTPS origin, native token apiKey:apiSecret authorization, one attempt, no redirects and no automatic retry. Workspace live execution must be enabled; WORKFLOW_LIVE_ERPNEXT supplies the deployment default. Existing finite DNS checks and TLS behavior remain unchanged. Local limits include 256 KiB of Fetch-exposed decoded body bytes, 1,024 reads including completion, JSON depth 16, 10,000 values/containers and 64 KiB of projected JSON. These and the name/token bounds can reject wider valid responses. Known retained compression headers do not cause another decompression or decoded-versus-wire length comparison. Provider 10-second and gateway 11-second execution budgets begin after confirmation and credential preparation; preparation itself has no deadline. Saving parameters, saving credentials or declining confirmation sends no request. Native v2 endpoints, triggers, credential Test and n8n import aliases are unsupported. Local synthetic tests, real provider behavior and hosted Review verification must be reported separately.

ERPNext document changes

Create one document, Update one document and Delete one document use the same selected native account and exact supported DocType/name rules. Create and Update require fieldsJson: a JSON object of 1–32 field names and string values, within 32 KiB. For example, {"description":"New task","priority":"0.10"} preserves decimal text without numeric conversion; {"description":""} can request clearing a field. The server validates required fields and permissions. Nested values, numeric/boolean/null values, duplicate keys, reserved system or routing-control fields (including cmd), ignore_* flags and docstatus transitions are unsupported. Update sends only the submitted fields; it does not fetch the existing document.

Privacy acknowledgment and document-change acknowledgment both start unchecked, remain editable and must be reviewed separately from run confirmation. Changes may trigger hooks, notifications or business workflows; deletion may be permanent or affect related records. Review the saved account, target and every submitted field. Local tests exercise an explicit simulated confirmation callback; the hosted approval path, actual human consent, permissions and durable effects remain unverified. A missing confirmation refuses execution.

The finite response contract follows Frappe v15.123.1 source: Create and Update accept HTTP 200 with supported document metadata; Delete accepts HTTP 202 with data: "ok" and returns accepted: true. This acknowledges the response and does not independently prove durable deletion. A different installed version may respond differently. No automatic retry, pre-read or verification request follows a write. If transport fails or the response is unsupported after dispatch, the change may already have occurred; inspect the target before deciding whether to retry manually. Provider authentication, version compatibility, hosted Review, organization scope and durable storage remain untested.

Freshworks CRM workflow nodes

Save an existing API key as freshworksCrmApi, using raw JSON with exactly two string members: {"apiKey":"your-existing-key","domain":"your-tenant"}. domain is one lowercase DNS label, up to 63 characters, for https://your-tenant.myfreshworks.com; URLs, ports, paths, custom domains and other Freshworks products are unsupported. The key is 1–2,048 visible ASCII characters without spaces or an added authentication prefix. These are local acceptance limits, not vendor field requirements. Saving sends no provider request and Test connection is unavailable; identity, privileges and expiry remain unverified. Settings validates after outer whitespace trimming. Other generic Save paths can retain malformed nonempty text unverified; execution validates the selected native account. Generic HTTP credentials, OAuth, names and Freshdesk/Freshservice accounts cannot substitute.

The 35 native operations cover six selected-record reads, six explicit single-page lists, search query and contact lookup, plus create/update/delete for accounts, contacts, deals, appointments, notes, sales activities and tasks. There are no import aliases or automatic conversion of legacy n8n parameters. This is a bounded field subset, not full node parity.

Review the exact operation and account. Personal-data acknowledgment starts unchecked and can be unchecked again. Writes also require a separate, initially unchecked effect acknowledgment for creation, changes, deletion, notifications or automation. Each invocation requires independent confirmation after these input acknowledgments; a programmatic callback returning true is not evidence of human consent. Existing durable approved-run confirmation remains limited to its supported Google operations; saving this CRM configuration does not grant durable approval. The existing explicit live single-step authorization is scoped to its checked workflow version and selected segment.

Whole expressions retain their boolean, number, array and string types; an expression returning undefined is rejected before confirmation and dispatch. Missing optional values remain absent. IDs are parsed from original JSON number tokens before numeric conversion, preserving values above JavaScript's safe integer range. A deal amount remains a decimal string. Local validation rejects unsupported nulls, unknown input keys and conditional combinations. Calendar/order checks and paired date updates are local policy. Appointment filters currently use open/complete; task filters are the listed supported values. Update omission, field clearing, attendee replacement, account-name-only updates, provider-wide validation and automation effects have not been verified against a real account.

Full responses can contain personal data and reach the bounded parser before scalar projection; this is not server-side field selection. Selected missing fields stay absent, and supported null, false, zero and empty strings stay distinct. Unknown fields, nested private content and unselected monetary response numbers are omitted. Decoded bodies are capped at 1 MiB, 4,096 reads including completion, depth 16 and 20,000 nodes; original numeric tokens at 128 characters, selected records at 100 and projected JSON at 16 KiB. Invalid or oversized responses fail without partial output. Compression validation uses the shared decoded-body policy, without an extra decompression or decoded-versus-wire length equality requirement.

Enable Freshworks CRM live execution for the workspace. A saved workspace boolean overrides the WORKFLOW_LIVE_FRESHWORKSCRM deployment default. Requests use the selected tenant's exact HTTPS origin, /crm/sales/api, and Authorization: Token token=<raw key>, without credentials or redirects. Every operation makes one attempt despite Retry On Fail; no retry, pre-read, returned-URL follow, automatic pagination or post-write verification occurs. Provider 10-second and gateway 11-second execution budgets begin after confirmation and credential preparation and do not bound that preparation. Writes accept the locally supported HTTP200/201 response shapes, and update/delete HTTP200; these status choices are not live-provider certification. Failures can leave a mutation outcome indeterminate.

Local tests use synthetic accounts, responses and explicit callbacks. Real provider authentication, product/domain compatibility, permissions, success/error shapes, mutation effects, hosted Review, organization authorization, durable persistence and hosted approval remain untested. Existing remote workflows are preserved.

Ghost workflow nodes

Use separate native ghostContentApi and ghostAdminApi accounts. Enter raw JSON with exactly two string members: {"adminOrigin":"https://your-site.ghost.io","apiKey":"your-existing-key"}. The origin must be a canonical lowercase HTTPS GhostPro root with one tenant label and no port, trailing slash, path, query or fragment. This GhostPro restriction is local support policy, not a vendor ban on self hosting. Duplicate or unknown members are rejected. Content keys use 1–1,024 visible ASCII characters. An Admin integration key uses a 1–128 character ID, a colon and exactly 64 hexadecimal secret characters. The complete JSON is locally limited to 4,096 UTF-8 bytes and validated original material is preserved after outer whitespace trimming. Do not substitute staff/password/session, generic HTTP, OAuth, a generic Ghost type or the other native profile.

Save account sends no request and creates no JWT; authentication, installed version, plan, key class and privileges remain unverified. Test connection is unavailable. Generic persistence paths can retain malformed nonempty text unverified; execution validates the selected account. Select that saved native account in the block and enable its separate Ghost Content or Ghost Admin workspace live control. WORKFLOW_LIVE_GHOST_CONTENT and WORKFLOW_LIVE_GHOST_ADMIN supply deployment defaults; an explicit workspace boolean overrides its default.

IDs are exactly 24 lowercase hexadecimal characters; slugs use 1–191 lowercase letters/digits with single hyphens. These and page/field limits are local narrowing. Each of the seven operations requires initially unchecked personal-data acknowledgment and separate invocation confirmation. The three writes also require initially unchecked effect acknowledgment. Unchecking persists false; saving configuration is not confirmation. Existing durable approved-run confirmation remains limited to supported Google operations. Explicit live single-step authorization is bound to its checked workflow version and selected segment; synthetic callback tests do not prove human consent.

Create requires an explicitly supplied HTML field; intentionally empty HTML remains empty. On Update, untouched HTML is omitted, while editing then clearing sends empty text. featured: false remains false. Whole expressions retain boolean, number and string types. An explicitly undefined value is rejected rather than converted into omission. Whole text expressions must resolve to strings: missing, null, boolean or numeric HTML never becomes a clearing write. Use an explicit empty string to clear HTML. Title is locally limited to 1–512 UTF-8 bytes, HTML to 65,536 bytes and a complete write input to 70,000 bytes. Updating uses the reviewed exact timestamp for concurrency; there is no automatic conflict fetch or retry. Scheduling requires a valid future RFC3339 date-time. Publishing, scheduling and deletion can affect public content, mail, webhooks and automation; deletion can be permanent.

Requests use unversioned /ghost/api/content/ or /ghost/api/admin/ routes with Accept-Version: v6.0, not the older archived n8n v2/v3 routes. Content sends its selected key only in the query. Admin creates a request-local HS256 JWT after authorization, confirmation and credential preparation, with the decoded 32-byte secret, selected key ID, /admin/ audience and a five-minute lifetime. Equal same-clock tokens can have equal bytes. API version negotiation can itself notify the site owner/admin; GET is not certified harmless. Key shape and a version header do not prove current server compatibility or permission.

Selected metadata requires an exact string ID and allows supported optional slug, title, featured, updated_at and published_at; Admin additionally permits a supported status. Missing fields stay absent; permitted null, false and empty text remain distinct. HTML, nested content, image URLs and raw diagnostics are excluded. Even with requested fields, full response content reaches the bounded local parser before projection. Unknown bounded fields are scanned for known secret collisions and discarded; collisions fail without partial output. No assertion of complete secret detection is made.

Local response limits are 256 KiB of Fetch-exposed decoded bytes, 4,096 reads including completion, depth 32, 20,000 JSON nodes, 10,000 object members and 32 KiB of projected JSON. Duplicate decoded keys and prototype-sensitive keys fail. Known retained compression headers do not trigger a second decompression or a decoded-versus-wire length equality check. Each operation makes one attempt even with Retry On Fail/maxTries 3, uses the exact selected origin, omits browser credentials and follows no redirects. Provider 10-second and gateway 11-second execution budgets begin after preparation; the two-second parser budget is cooperative and preparation itself has no deadline.

After a dispatched write fails, times out or returns an unsupported response, the write outcome is uncertain. Inspect the selected site before a deliberate manual retry; no pre-read, verification call or automatic replay follows. No triggers, automatic pages, lookup loaders, author/tag/newsletter controls, uploads, n8n import aliases, Pro/anonymous fallback or OAuth refresh are added. Local tests use synthetic responses and credentials. Real provider authentication, plan/key privileges, deployed version, owner mail, DNS/TLS, hosted Review, organization authorization, durable storage and human approval remain untested.

Gotify workflow nodes

Save a native gotifyApi account as raw JSON with url and appApiToken, clientApiToken or both. Example: {"url":"https://push.company.com","appApiToken":"existing-application-token","clientApiToken":"existing-client-token"}. The origin must be a canonical lowercase public root HTTPS origin without a trailing slash, explicit port, path, query or fragment. Private/local/IP hosts and self-signed TLS are outside this local support profile; self hosting on a supported public origin remains possible. An optional ignoreSSLIssues member must be false. Each supplied nonempty token uses 1–4,096 visible ASCII characters; these are local support bounds, not a vendor token format. Empty unavailable roles are omitted, but at least one role is required. Received raw JSON is limited to 32,768 JavaScript string units before trimming; duplicate decoded or unknown keys are rejected.

Save sends no provider request and leaves identity, privileges and server version unverified. Test connection is unavailable. Settings validates native JSON locally and preserves original validated material after outer whitespace trimming. Lower generic persistence can store malformed nonempty text unverified; execution validates the exact selected saved native account. Generic HTTP/OAuth credentials, credential names and another workspace's accounts cannot substitute. Enable the Gotify workspace live control; an explicit workspace boolean overrides WORKFLOW_LIVE_GOTIFY.

Both supplied token roles are retained and never substituted. Optional title intentionally accepts an empty string; omission stays absent and priority zero stays zero. Priority accepts safe integer numbers, excluding negative zero, or canonical signed-int64 decimal strings; exact strings preserve larger values. Whole text expressions must return strings. Missing, null, boolean or numeric title expressions are refused before confirmation and dispatch. IDs/cursors require strings. No n8n import aliases or automatic parameter conversion are supplied. Reconfigure imported nodes manually and review every field; no automatic pagination or full-node parity is provided.

Privacy acknowledgment starts unchecked on all three operations. Create and Delete also require a separate initially unchecked effect acknowledgment. These remain editable, unchecking persists false and each invocation requires independent confirmation. Creating can notify devices or people; deletion can permanently remove messages. A callback returning true in a local test does not demonstrate human consent or durable approval. Live single-step authorization is checked against its current workflow version and selected segment; hosted ordinary approval behavior remains unverified here.

Full upstream messages can contain personal data and reach the bounded parser before selection. Message text, titles, extras and raw diagnostics are not returned. Original numeric integer lexemes preserve id/appid/priority/count precision. Missing optional values remain absent; supported null and empty paging markers remain distinct. Returned next values are not emitted as URLs or followed; hasMore only reflects the marker and is not a completeness guarantee.

Requests use the selected exact origin, role-specific X-Gotify-Key header and fixed /message routes, omit browser credentials and follow no redirects. One attempt applies despite Retry On Fail/maxTries3; there is no retry, pre-read, post-write verification or automatic page. Provider10-second/gateway11-second budgets begin after confirmation and credential preparation; preparation has no deadline. Local response bounds include256KiB of Fetch-exposed decoded bytes,1,024 reads including completion, depth32,20,000 JSON nodes,40,000 tokens,65,536 characters per JSON string and64KiB of projected JSON. Some later caps are shadowed by earlier bounds. Known retained compression headers use the shared policy without second decompression or decoded-versus-wire length equality. These limits can reject valid wider server responses.

After a dispatched write fails, times out or returns an unsupported response, the change may already have occurred. Inspect the selected server before deciding on a manual retry. The finite contract follows shipped n8n2.7.1 operation evidence and the separately pinned Gotify v3.1.1 API schema. Its42 documented API operations include39 outside this subset; this is partial coverage, not proof those operations are impossible. No trigger, login, token issuance, application/client administration, server discovery, arbitrary URL, WebSocket or OAuth feature is added. Local synthetic app/UI tests do not verify real provider authentication, permissions, version compatibility, DNS/TLS, notification/deletion effects, hosted Review, organization authorization, durable persistence or human approval.

Hunter workflow nodes

Save an existing hunterApi key as raw text, then select that exact native saved account and enable Hunter live execution. Save sends no request and leaves identity and permissions unverified; Test connection is unavailable. Generic HTTP/OAuth accounts and names cannot substitute for the selected native account. Settings trims outer whitespace before local validation; native execution accepts 1–2,048 visible ASCII characters and a separately selected account ID.

Search email addresses by domain requires a lowercase domain, an explicit numeric limit from 1 to 100 and personal-data acknowledgment. Optional offset is an integer from 0 to 10,000,000; type is personal or generic. Optional department and seniority filters contain 1–16 distinct strings without commas. One page returns selected email/name/confidence fields and available page metadata; it does not prove completeness.

Find an email address requires domain, first name and last name. Verify requires an email address. Both also require explicit acknowledgment of credit use and automatic lead saving or updating, even though requests use GET. Personal-data and effect switches start unchecked, remain editable and do not replace separate invocation confirmation. Whole expressions retain original supported types, including rejecting a null type filter rather than taking an earlier node's value.

Requests use the selected key only in X-API-KEY at the fixed HTTPS Hunter v2 origin. No environment fallback, redirect, automatic retry, page loop or follow-up verification is performed, including when Retry On Fail is enabled. Structured inputs, the final composed URL and selected output strings are checked against known key fragments; a colliding key or input is refused. Provider and gateway budgets are 10 and 11 seconds after confirmation and credential preparation. Preparation itself has no deadline. Fetch-exposed responses are limited to 1 MiB and 4,096 reads, with bounded JSON depth, nodes, strings and selected output.

A supported HTTP 202 verification envelope returns pending without polling or a fabricated score; HTTP 222 is a terminal refusal locally. A failed or timed-out request may already have consumed credits or changed leads. Inspect Hunter before deciding on a manual retry. These three shipped n8n operations are a subset of the vendor API; other operations remain backlog. Local synthetic tests do not verify provider credentials, plan, authorization, credits, lead persistence, hosted Review, organization scope or human consent.

Home Assistant workflow reads

Save homeAssistantApi as raw JSON containing exactly apiOrigin and accessToken, then select that saved native account. The local profile accepts a public root HTTPS origin on port 443, without credentials, path, query or fragment; local/private/IP and punycode hosts are refused. An existing token must contain 16–8,192 supported ASCII characters; the original JSON is capped at 16,384 UTF-8 bytes. These are local support limits, not vendor token specifications. Settings trims outer whitespace and validates locally. Save sends no provider request and leaves authentication, ownership, permissions and version compatibility unverified. Test connection is unavailable and preserves existing credential-read restrictions; generic nonempty secret storage can still hold malformed unverified material.

Enable Home Assistant live execution in workspace settings. The explicit workspace setting takes precedence over the WORKFLOW_LIVE_HOME_ASSISTANT environment fallback. Select Get state, List states, Get configuration, List event listeners or List service names. Get state requires an exact lowercase domain.object_id entity ID. Personal-data acknowledgment starts unchecked, remains editable and is separate from invocation confirmation. Manual confirmation is tied to the checked workflow version and segment; unattended and durable continuation are not certified by local tests.

Each read uses the selected account's exact origin and Bearer token with one GET, credentials omitted and redirects refused. There are no device actions, service calls, event firing, writes, automatic retries, pagination, polling, discovery, OAuth refresh, n8n import aliases or parameter conversion. Retry On Fail does not add attempts. Provider and gateway budgets are 10 and 11 seconds after confirmation and credential preparation; preparation itself has no deadline. A timeout does not establish freshness or server compatibility.

Only selected state strings/timestamps, configuration version/state/time zone/components/four unit names, event names/listener counts and domain/service names are returned. Household attributes, coordinates, location names, contexts and service descriptions are omitted. The full received JSON still reaches bounded UTF-8 and duplicate-key parsing before projection. Known credential material is checked in actual selected object keys and string values; array values are scanned, numeric/boolean/null values are not coerced, and discarded private fields are not scanned as public output. Missing, null, empty strings and zero retain the meanings permitted by each operation. Listener counts require original unsigned integer JSON spellings within the safe-integer range.

Fetch-exposed decoded bytes are capped at 512 KiB and 4,096 reads including empty and terminal reads; supported coding labels follow the shared five-coding limit without a second decompression or wire-length equality assumption. Parsing and output have separate depth, node, string and size bounds; limits refuse the entire result. Lists accept at most 250 states, 256 event names or 128 domains, with at most 128 services per domain and 2,048 services total. These five reads cover part of the shipped node and do not claim full node or vendor API parity. Local synthetic tests do not establish real-provider authentication, DNS/TLS, hosted Review, organization authority, durable persistence or human consent.

Humantic AI profile operations

Save humanticAiApi with an existing API key, then select that exact native account in the block. Advanced JSON must contain exactly apiKey. The received JSON limit is 32,768 JavaScript string units, including outer whitespace; the key accepts 1–4,096 visible ASCII characters without trimming. These are local support limits, not a vendor key specification. Settings validates the original received JSON locally, preserves it after outer whitespace trimming and sends no provider request. Saving remains unverified and Test connection is unavailable; existing credential-read authorization still applies. Lower generic nonempty storage can retain malformed unverified material, which execution refuses.

Enable Humantic AI live execution in workspace settings. The explicit workspace value overrides WORKFLOW_LIVE_HUMANTICAI. Both actions require initially unchecked, editable personal-data and processing/credits acknowledgments plus separate invocation confirmation. Acknowledgment and run confirmation do not establish the person's consent. Creation accepts a supported LinkedIn profile URL or email address; metadata retrieval also accepts an opaque ID. IDs are limited to visible ASCII and 2,048 characters, with a narrower local email profile. For metadata, optionally provide persona as a JSON array containing unique sales and/or hiring values; [] is an explicit empty array. Whole expressions preserve their original types, so booleans must resolve to actual booleans.

Create profile request starts processing and is classified as a write even though Humantic uses GET /v1/user-profile/create. Email creation may use the provider's default social enrichment. Get profile metadata uses GET /v1/user-profile and can consume credits. Requests use fixed https://api.humantic.ai routes with id and a query apikey, plus persona when supplied. The API key is intentionally present in the provider request URL; it is not sent in an Authorization header. The adapter checks the complete non-auth query for known secret fragments before appending that required key. Only selected metadata strings and original numeric spellings are exposed; personality assessments and other personal details are discarded after the full bounded response arrives. Returned user IDs are provider reports, not verified identities. Missing fields, null, code zero and valid replacement characters keep their supported meanings. Numeric integer IDs preserve original digits without rounding.

Every invocation makes at most one request even with Retry On Fail. Redirects, automatic retries, polling, updates, text/file uploads, n8n import aliases and parameter conversion are unavailable. Provider and gateway budgets are 10 and 11 seconds after confirmation and credential preparation; preparation itself has no deadline. An aborted, failed or timed-out creation can already have started processing: inspect the provider before deciding to create another request. No completion, freshness, quality, credit balance or durable outcome is established.

The adapter requires its exact nonempty response URL, HTTP 200 and the operation's supported JSON status/envelope. Fetch-exposed decoded data is capped at 1 MiB and 4,096 reads, with supported content-coding labels following the shared five-coding limit; no second decompression or wire Content-Length equality is assumed. Original JSON rejects duplicate keys and has separate depth16, node20,000, string, number and output limits. Known raw/URI/form credential fragments are checked against actual selected object keys and string values, including array values without scalar coercion. A bound, unsupported shape or secret collision refuses the entire result with a static error. These two operations cover only part of the shipped node. Real provider authentication, DNS/TLS, hosted Review, organization authority, durable persistence and human consent remain untested in local synthetic checks.

Iterable events, profiles and subscriptions

Save iterableApi with an existing API key and an explicit project region, us or eu; neither is inferred. Advanced JSON contains exactly apiKey and region. The original received JSON, including outer whitespace, is limited to 32,768 JavaScript units; keys accept 1–2,048 visible ASCII characters without internal trimming. Settings validates locally and trims only outer JSON whitespace. Save sends no provider request, remains unverified and offers no Test connection. Select the saved native account and enable Iterable live execution in workspace settings; an explicit workspace value overrides WORKFLOW_LIVE_ITERABLE.

Six operations cover event tracking, user upsert, selected user fields, user deletion, list subscription and list unsubscription. All require the initially unchecked personal-data acknowledgment and separate invocation confirmation. Five writes also require the initially unchecked effects acknowledgment. Setting channelUnsubscribe to true requires its own channel acknowledgment because it can unsubscribe the list-associated channel globally. Switches remain editable. Accepted writes do not certify completion, consent or durable effects; event/profile/list changes can trigger automation, and deletion is asynchronous rather than GDPR forgetting.

Use exact string identifiers and explicit JSON arrays. List, campaign and template IDs are positive decimal strings through 9223372036854775807; event epoch seconds are decimal strings from 0 through 253402300799. The adapter emits their original digits as JSON numbers without binary rounding. Whole expressions retain their types. Upsert accepts 1–32 flat string data fields, including empty values, and an explicit mergeNestedObjects boolean. preferUserId is supported only with a userId identity. User reads require 1–32 unique literal field names and return selected scalar fields plus missing-field names. Subscription batches accept 1–25 unique identifiers. These are local support limits.

The selected region fixes https://api.iterable.com or https://api.eu.iterable.com; requests use Api-Key without region fallback or the historical Api_Key spelling. Each invocation makes one attempt, including with Retry On Fail, without redirects, verification reads, polling or automatic pages. Provider and gateway budgets are 10 and 11 seconds after confirmation and credential preparation; preparation has no deadline. A failed or timed-out write can already have applied: inspect Iterable before choosing a manual retry.

Full private responses reach bounded parsing before projection. Fetch-exposed decoded data is limited to 1 MiB and 4,096 reads; JSON depth16, nodes20,000 and selected output64KiB have separate limits. Supported content-coding labels follow the shared five-coding policy without a second decompression or wire-length equality assumption. Original JSON rejects duplicate keys, malformed UTF-8 and lone surrogates. Selected permitted null, false, zero, empty and missing values stay distinct; valid replacement characters in output are retained. Known credential fragments are checked in final request URL/body string material and actual selected keys/strings. A limit or collision refuses the whole result with static diagnostics. The parameter editor currently normalizes duplicate JSON keys before provider validation; duplicate rejection therefore applies to the provider's received object/response boundary, not original editor text.

These six operations are a subset, with no n8n import aliases or parameter conversion. Local synthetic tests exercise Settings, selected accounts, manual routing, controls, retry and cancellation; they do not establish real provider authentication, DNS/TLS, authorization, hosted Review, organization scope, durable persistence or human consent.

HighLevel legacy PIT operations

Save highLevelPitApi using an existing sub-account private integration token (PIT), the intended location ID and the explicit legacy API version 2021-07-28. This native profile is separate from OAuth and API v1; it does not acquire, refresh or substitute credentials. Advanced JSON contains exactly accessToken, locationId and apiVersion. Settings validates the original received JSON before trimming outer whitespace, with a local 32,768 JavaScript-unit limit. Tokens accept 1–4,096 visible ASCII characters and location IDs accept 1–128 ASCII letters, digits, underscores or hyphens. These are local support limits, not vendor token specifications. Save sends no provider request and remains unverified; Test connection is unavailable and existing credential-read authorization still applies. Generic nonempty storage may retain malformed material that execution refuses.

Select the exact saved native account and enable HighLevel workspace live execution. The explicit workspace value overrides WORKFLOW_LIVE_HIGHLEVEL. All seventeen operations require the initially unchecked, editable personal-data acknowledgment and separate invocation confirmation. Ten writes also require the initially unchecked effects acknowledgment; contact upsert additionally requires acknowledgment that an existing contact may be modified. These switches and a local manual grant do not establish human consent, account ownership, permissions or durable completion.

The seven reads are contact-get, task-get, contacts-list, free-slots, opportunity-get, task-list and opportunity-list. Writes are contact-delete, appointment-create, contact-upsert, contact-update, opportunity-create, opportunity-update, task-create, task-update, opportunity-delete and task-delete. Appointment creation can book a real meeting; fixed notification flags do not prove that every downstream effect is suppressed. Contact/task/opportunity changes can trigger automation or deletion. Upsert's exact required permission scope remains unresolved locally: use only a correctly authorized PIT. A matching response location is checked when present; missing location does not prove membership.

Keep opaque IDs as exact strings, up to 128 supported ASCII characters. Contacts-list requires numeric limit1–20 and optionally preserves an original query up to512 JavaScript units/512 UTF-8 bytes. Opportunity-list requires numeric limit1–20, makes one legacy page1 request and accepts optional q up to75 units/300 UTF-8 bytes. Task-list rejects a response with more than20 tasks; no complete-list guarantee or invented paging parameter is supplied. Free-slot dates are original13-digit millisecond strings with a positive interval of at most7days. Booking requires explicit future ISO times within31days and a duration of at most1day. Tasks use explicit ISO due dates and actual boolean completed values; false is retained. Create requires nonempty names/titles; updates send only the supported explicit fields.

For supported optional query or contact-name fields, edit and clear the control to send an explicit empty string. Omitted names stay unsent. To clear a nullable contact name, use a whole typed expression such as {{input.clear}} whose value is null, or {{ null }}. Typing null in a literal textbox sends text, and the inspector does not distinctly round-trip manually authored top-level nulls. Keep credentials out of block fields. Failure diagnostics mask credentials already resolved in that execution, including their supported encoded forms; this does not erase original saved workflow inputs or earlier history. Provider refusal prevents dispatch. Whole expressions preserve original scalar types, including rejecting strings in boolean/numeric fields. Returned permitted null, false, zero, empty and missing values retain their meanings.

Requests use Bearer authentication and the fixed Version header at https://services.leadconnectorhq.com. Only fixed supported routes are composed; redirects, returned-URL following, discovery, automatic pagination and retries are unavailable. Each invocation makes one attempt even with Retry On Fail. Provider10s/gateway11s budgets begin after confirmation and credential preparation; preparation has no deadline. A failed or timed-out write may already have applied: inspect HighLevel before choosing a manual retry.

Full private responses reach bounded parsing before selected projection. Fetch-exposed decoded data is capped at1MiB and4096reads, with the shared five-coding policy and no second decompression or wire-length equality assumption. Original JSON rejects duplicate keys and has separate depth16, node10000, string, numeric-token and selected-output32768-byte UTF-8 bounds. Selected scalar metadata is returned rather than whole contacts, tasks, opportunities or arbitrary provider bodies; task-get includes its explicitly selected task text. Opportunity money, nested contact details, page links and aggregations are omitted. Original safe-integer count spellings are checked before conversion; numeric IDs are refused. A bound, unsupported shape or known-secret collision rejects the whole result with static diagnostics.

This finite legacy subset does not claim complete shipped-node or current vendor API parity. Task body/assignee writes, pipeline/stage reads, triggers, n8n import aliases and automatic parameter conversion remain outside it. Local synthetic checks cover both workspaces, Settings, controls, selected accounts, persisted manual grants, one-attempt failures and representative late cancellation. Real provider authentication, DNS/TLS, scope availability, hosted Review, organization authority, durable persistence and human consent remain untested.

Keap bounded reads

Save a Keap personal access token as the native keapApi account, then select that saved account in the block. Advanced JSON contains exactly accessToken, for example {"accessToken":"your-existing-token"}. This profile is separate from OAuth; it does not acquire or refresh a token. Saving sends no provider request and remains unverified. Test connection is unavailable. A saved profile cannot establish the supplied token's actual type, account, permissions or expiry. Keep credentials in saved accounts, outside workflow input fields.

Settings validates the originally received JSON before outer whitespace trimming. Local support limits are 32,768 JavaScript units for the received JSON and 1–2,048 visible ASCII characters for the original token. These bounds describe this integration, not a vendor token format. The generic nonempty store can retain malformed material that execution rejects.

Enable Keap workspace live execution. Its explicit workspace setting overrides WORKFLOW_LIVE_KEAP. Every action requires the initially unchecked, editable personal-data acknowledgment and separate invocation confirmation. Writes also require the applicable effect acknowledgments; none starts checked. Names, note text, email subjects and file metadata can contain personal data. The response is received privately before selected fields are returned. These controls do not prove account ownership or human consent.

This profile supports 29 actions: twelve reads and seventeen writes. The twelve supported reads are List companies, Get contact, List contacts, Get note, List notes, List contact tags, Get order, List orders, Get product, List products, List email records and List file metadata. Every list requires a numeric limit from 1 to 20 and reads one first-page response. Companies require a company-name filter, contacts require an email filter, contact tags require a contact ID, products require an explicit active/inactive choice, and notes/orders/email records/files require a contact ID. Original filter text is preserved. Filters can match several records and results may be incomplete; there are no automatic pages, returned-link requests or ownership checks inferred from a filter. A response exceeding the requested limit is rejected, not truncated.

Keep IDs as original positive decimal strings, from "1" through "9223372036854775807", without spaces or leading zeros. Response numeric IDs retain their original decimal spelling instead of being rounded through JavaScript numbers. Get operations require the actual returned ID to match the requested ID. A contact response without its own numeric ID is rejected: the current contact schema omits that field, so this local compatibility rule can reject a provider-valid response. Notes and email records check the contact ID when it is present; absence proves no ownership.

Applied tags return selected fields from the nested tag, including its string category. File metadata returns remote_file_key, optional file name, public visibility and decimal-string file size, including "0". The descriptor key is not a file entity ID and cannot be used as a fabricated delete ID. No file content or download links are returned. Orders omit money, payment and customer detail; email records omit addresses, headers and content; products omit prices and plans. The current product listing is deprecated and returns no active field, so the requested filter is never copied into the result. Selected empty strings and false values are retained; missing fields stay absent, while unsupported nulls and types reject the response.

The seventeen writes are Create company, Create note, Update note, Apply contact tags, Remove contact tags, Upsert contact, Delete contact, Delete note, Delete order, Delete product, Delete email record, Delete file, Create order, Create product, Create email record, Queue email and Upload file. All use fixed routes and explicit fields; arbitrary record JSON is unavailable. Contact upsert can create or merge and requires an explicit duplicate rule. Its local response checks can reject a provider-valid response because the current contact schema omits the numeric ID. Tag changes require 1–20 unique exact IDs; applying tags preserves individual failure statuses without inventing aggregate success.

Deletion needs a separate removal/deletion acknowledgment. Order/product changes require financial acknowledgments, and product deletion additionally acknowledges subscription effects. Order creation requires an explicit contact, date, title, type and 1–10 strict line items, each with a product ID, numeric quantity1–1000 and original unitPrice decimal string: integer0–999999 and up to six fractional digits. Prices use the provider account currency; this integration does not identify it, charge a payment or project monetary totals. Product creation requires explicit false active and subscription-only flags and acknowledges provider financial defaults; it does not promise a free product.

Create email record stores history only; it does not send mail. Queue email requires explicit user/contact IDs, subject and canonical base64 content, with sending and recipient-effects acknowledgments. An empty202 proves transport acceptance only, without a job, queued, sent or delivered guarantee. File upload supports contact files only, with an explicit contact, name, canonical base64 and public/private flag. Storage acknowledgment is required; public files need another distribution acknowledgment. No content or download URL is returned. Decoded text content is limited to8,192 UTF-8 bytes and uploaded files to32,768 bytes; request JSON is capped at65,536 UTF-8 bytes. Successful deletion/removal requires an empty204 body and returns no invented entity.

Requests use fixed HTTPS Bearer-authenticated RESTv1 routes at https://api.infusionsoft.com/crm/rest/v1. One attempt applies even with Retry On Fail. Redirects, retries, alternate hosts, arbitrary headers and follow-up requests are unavailable. The provider's 10-second and gateway's 11-second execution budgets start after confirmation and credential preparation; preparation has no deadline. A slow or failed read returns a static error, without provider diagnostics. After a write request begins, a failure can mean the change took effect: inspect the provider before an explicit manual retry. Error text can be reduced further when it collides with a known saved secret.

Local decoding and parsing accept at most 256 KiB of fetch-exposed decoded data, 4,096 stream reads, depth32, 10,000 JSON nodes and 32,768 JavaScript units per raw decoded string. Selected ordinary strings allow512 units/2,048 UTF-8 bytes, note text2,048 units/8,192 bytes and descriptor keys1–256 units/1,024 bytes. Public output is capped at16 KiB. Invalid UTF-8, duplicate decoded keys, malformed Unicode, unsupported response shapes or a bound violation reject the whole response. Accepted content coding describes fetch-exposed data and is not decompressed twice.

This subset does not add triggers, dynamic account lookups, OAuth renewal, n8n import aliases or automatic parameter conversion. Reconfigure imported nodes using the supported fields. Current PAT permissions, server query serialization, deprecated endpoint availability and responses remain dependent on Keap; saving an account does not verify them. Local synthetic application checks do not certify real-provider authentication, DNS/TLS, hosted Review, organization authority, durable execution or human approval.

Invoice Ninja hosted v5

Select a saved native invoiceNinjaV5Api account for the hosted invoicing.co v5 service. Advanced JSON contains exactly apiToken, for example {"apiToken":"your-existing-token"}. This profile does not support self-hosted servers, v4, OAuth, arbitrary API URLs or token acquisition. Saving sends no provider request and remains unverified; Test connection is unavailable. Identity, permissions, token type and expiry are not established by a saved profile.

Settings validates the originally received JSON before outer trimming, with a local limit of 32,768 JavaScript units. The original token must contain 1–4,096 visible ASCII characters. These are local support bounds, not a vendor token format. Generic nonempty persistence can retain malformed material that native execution refuses. Keep credentials in saved accounts, outside block inputs.

Enable Invoice Ninja workspace live execution and select the exact account in each block. An explicit workspace choice overrides WORKFLOW_LIVE_INVOICENINJA. Every operation requires the editable personal-data acknowledgment, initially unchecked, and independent invocation confirmation. Creation, deletion, financial changes, linked-record effects and email actions require the additional acknowledgment switches shown for that operation. A simulated confirmation callback or local test is not human consent.

This finite profile provides 31 operations:

Keep record IDs as exact strings of 1–128 ASCII letters, digits, underscores or hyphens. Client Get uses clientId; other Get operations use id. Lists require numeric page from 1 to 1,000 and numeric perPage from 1 to 25. They return one partial page, never a complete listing or automatic continuation. Returned links are inert; an oversized page is rejected rather than silently truncated. An ID correlation checks the actual returned ID, not account ownership.

Client Create sends the supplied name and an empty contacts array. Task Create sends the description and an empty time log. These operations do not create contacts or start a task timer. Bank-transaction Match takes one transaction ID and one payment ID; it performs no lookup to verify their business relationship. Quote Email uses a historical GET route but remains an effect that requires acknowledgment and confirmation. Its message does not prove delivery.

Invoice and Quote Create require an existing client ID, a Gregorian YYYY-MM-DD date and 1–25 line items. Each item contains productKey (1–128 characters), notes (0–2,048), unitCost and quantity. Use the collection editor or an expression returning an actual strict array of those objects; do not supply a quoted JSON string as a typed array. Empty notes remain empty. Whole expressions preserve original scalar types and business IDs remain strings.

Financial amounts, costs and quantities are original decimal strings, with 1–12 integer digits and an optional 1–6 fractional digits, no signs, exponent or leading zeros except zero itself. Unit cost, expense amount and bank amount may be zero; quantity and payment amount must be greater than zero. Original spelling is serialized directly as a JSON number without passing through JavaScript numeric rounding. This profile does not calculate totals, convert currencies or establish the provider's currency scale.

Payment Create requires client ID, invoice ID, date, positive amount and a visible-ASCII idempotencyKey of 1–64 characters. It allocates that amount to the one selected invoice and explicitly requests no receipt email. The field does not establish a cross-request idempotency or payment-settlement guarantee. Expense Create requires date, currency ID and amount and explicitly disables invoicing. Bank-transaction Create requires bank-integration ID, currency ID, date, amount, explicit DEBIT or CREDIT, and a description that may be empty. Currency IDs are positive decimal strings of at most nine digits, not currency codes. Provider defaults and linked financial effects still need review.

Deletes send the exact selected resource ID and require the actual response identity and deletion flag. They do not independently verify downstream accounting effects or durable deletion. Request invoice email sends the selected ID in the bulk-email request; its output always reports emailOutcome: "unverified". Neither accepted status nor an email action proves a message was sent, delivered or read.

Requests use fixed HTTPS routes under https://invoicing.co/api/v1, raw X-API-TOKEN authentication and X-Requested-With: XMLHttpRequest. Only the supported HTTP 200 response shape is accepted. There is one attempt even with Retry On Fail, no redirects, arbitrary headers, account discovery, automatic pages or verification reads. Provider and gateway budgets are 10 and 11 seconds after separate confirmation and credential preparation; preparation has no deadline. A dispatched effect can remain indeterminate after timeout, cancellation, a rejected response or publication failure. Inspect provider state before a deliberate new action.

Responses are received privately before selected scalar projection. Local limits are 256 KiB of fetch-exposed decoded data, 4,096 stream reads, depth 64, 20,000 JSON nodes, 32,768 units per raw string token and 64 KiB of public JSON output. Invalid UTF-8, duplicate decoded object keys, malformed Unicode, unsupported shapes and limit violations reject the response. Fetch-exposed data is not decompressed twice. Supported missing/null/false/zero/empty values remain distinct. Selected numeric financial response tokens remain bounded original text, including valid exponent spelling; this is not an IEEE finite-value or arithmetic guarantee. Private nested customer details are not dumped into output.

Queued workflows containing enabled native API blocks reserve a segment before execution. A duplicate or uncertain claimed segment is refused; an expired worker lease does not permit blind native replay. Preparation can fail after reservation without any HTTP request, so inspect the run and provider state before deciding whether to start a new run. There is no automatic claim-reset or reconciliation control. An actual committed wait/resume admits its exact continuation without rerunning earlier blocks; this is not an external exactly-once guarantee. See native execution and recovery.

Other vendor operations, self-hosted/v4 profiles, contact enrollment, triggers, file downloads, automatic n8n parameter conversion and import aliases remain outside this subset. Local synthetic transport and actual disposable-Postgres checks do not establish provider authentication, DNS/TLS, hosted Review, organization-wide behavior, hosted BullMQ delivery or human approval.

LinkedIn confirmed publication

Save an existing OAuth access token as the native LinkedIn existing OAuth access token account (linkedInAccessTokenApi), then select that saved account in the block. Advanced JSON contains exactly accessToken, for example {"accessToken":"your-existing-token"}. This manual profile does not acquire, refresh or test OAuth tokens and does not accept historical n8n OAuth credentials automatically. Save sends no provider request and remains unverified. Settings validates the original received JSON before outer trimming: 32,768 JavaScript units, with 1–4,096 visible ASCII characters for the original token. These are local support bounds, not a vendor token format. The generic nonempty store can retain malformed material that execution rejects. App approval, scopes, author role, identity and expiry remain unverified.

Enable the separate LinkedIn workspace live-execution control; an explicit workspace choice overrides WORKFLOW_LIVE_LINKEDIN. Choose Create text post, Create article post or Create post with existing image. Each requires explicit authorUrn, visibility, literal text, and the initially unchecked, editable acknowledgePublication switch. Review personal data, author, visibility and content before checking it, then confirm the exact invocation separately. A confirmation callback in a test is simulated, not evidence of human consent. Publication can expose content publicly and repeated requests can create duplicate posts.

Use an explicit person URN (urn:li:person:…) or organization URN (urn:li:organization:…); no identity lookup chooses one. Visibility is PUBLIC or CONNECTIONS, with organizations restricted locally to PUBLIC. Text is nonblank, well-formed Unicode of at most 1,500 JavaScript units; tab and line breaks are supported. Mention, hashtag and formatting punctuation is escaped as literal text. Article posts require sourceUrl, title, description and a preexisting thumbnailImageUrn; requiring all four is a local narrowing. The HTTPS source URL is data and is not fetched. Image posts require a preexisting imageUrn and title. Titles allow at most 200 units, descriptions 500 and source URLs 2,048. Existing image URNs do not prove ownership or readiness. No binary upload, media preparation, post reading, listing, deletion, targeting or trigger is included. LINE Notify is retired and unsupported.

Whole expressions preserve original scalar types. acknowledgePublication must evaluate to the boolean true, not the string "true". Keep account selection outside tool inputs. Requests use the fixed LinkedIn Posts endpoint and explicit API version 202610; that choice is not proof of current app access or version acceptance. Member publication needs the appropriate member permission; organization publication also needs its organization permission, author role and app approval. These external prerequisites have not been tested.

One request is attempted without retry, readback or fallback. The provider budget is ten seconds after credential and local preparation; the shared gateway budget is eleven seconds after credential preparation. Preparation has no deadline. Only an actual 201 response with the exact fixed URL and valid x-restli-id header returns {urn}. The original decimal URN spelling is preserved, including leading zeros; no body, success flag or timestamp is invented. Response bodies are never read and an exposed body is canceled. A returned URN does not verify public visibility or media readiness. After an attempted request, failures are indeterminate; inspect LinkedIn before explicitly deciding whether to run again. Stop cannot undo a remotely completed publication.

Local synthetic application checks cover saved A/B accounts in two workspaces, persisted manual grants, native controls, strict Settings validation, the real server guard and one-attempt handling. They do not certify real credentials, LinkedIn permissions, delivery, matching-revision Review, organization authorization or human consent. The preexisting LinkedIn brand icon is reused. This finite profile is partial n8n service coverage and adds no automatic n8n parameter or authentication conversion.

Mandrill Transactional email

Use the native mandrillTransactionalV14Api account with an existing Mailchimp Transactional API key, then select that saved account in the block. Advanced JSON contains exactly apiKey, for example {"apiKey":"your-existing-key"}. This profile is separate from the historical n8n mandrillApi credential and has no automatic import alias. Saving sends no provider request and stays unverified; Test connection is unavailable. Settings validates the originally received JSON before outer whitespace trimming, with local limits of 32,768 JavaScript units and an original visible-ASCII key of 1–2,048 characters. These limits do not identify a vendor token format. The generic nonempty store can retain malformed material that runtime rejects. Live or test key status, account identity, permissions, delivery and billing remain unverified.

Enable the Mandrill workspace live-execution control; an explicit workspace choice overrides WORKFLOW_LIVE_MANDRILL. Choose Send HTML message or Send template message. Both require explicit fromEmail, toEmails, subject and asyncMode, plus the initially unchecked, editable acknowledgePersonalData, acknowledgeEmailSend and acknowledgeBillableSend switches. Review recipient and content data, email effects and possible charges, then confirm the exact invocation independently. Simulated confirmation callbacks are not evidence of human consent. A reported sent status can come from a test key and is not delivery proof.

Recipients are an explicit array of 1–10 unique original ASCII mailboxes. The local mailbox grammar and 254-character limit do not verify a real address or authorized sender. The original sender/recipient spelling is preserved. Subject is nonblank, at most 256 JavaScript units/1,024 UTF-8 bytes, with no line breaks or NUL. HTML requires a nonblank html string, at most 8,192 units/32,768 UTF-8 bytes, and offers optional text with the same size bounds. Untouched optional text stays absent; explicitly edited empty text stays empty after Save/reload. Whole expressions preserve the original string type, including empty text; false, null, numbers and objects are rejected. Content is sent as supplied; this profile does not render HTML in output.

Template sends require an existing immutable templateSlug, explicit templateContent, globalMergeVars and mergeLanguage (mailchimp or handlebars). The two variable arrays each allow 0–20 strict {name,content} objects. Names are nonblank, at most 64 units/128 UTF-8 bytes, cannot start with an underscore and must be unique ignoring case within each array. Content can be empty, with a 2,048-unit/8,192-byte limit. Empty arrays stay arrays. The template slug allows 128 units/512 bytes; template content and vendor permissions are not read or verified. All acknowledgments must evaluate to actual boolean true; asyncMode can be actual false. Keep credentials and selected-account context outside message inputs.

The fixed API v1.4 profile makes one POST to the HTML or template send route, carrying the selected API key in the JSON body. Tracking, recipient exposure and public view-content links are explicitly disabled; this local setting is not a broader vendor privacy guarantee. There are no arbitrary routes/headers, redirects, attachments, template creation, scheduling controls, polling, delivery webhooks, readback or automatic retries. The request is capped at 64 KiB UTF-8.

Only an actual HTTP 200 response with the exact fixed response URL, supported JSON media/decoded content coding and a complete bounded JSON array is projected. At most ten selected recipient records return original email, reported status, actual string id and optional finite rejectReason/queuedReason. An actual empty ID is preserved; no ID, delivery flag or timestamp is invented. Results must correlate uniquely to the exact requested recipients; response order is retained. Supported statuses are sent, queued, scheduled, rejected and invalid. Present null or unrecognized reason values are rejected as a local restriction. Unknown provider fields are discarded and no raw recipient/content dump is returned.

The exposed decoded response is capped at 128 KiB, with fatal UTF-8 decoding, 4,096 read iterations, original decoded duplicate-key rejection, depth 32, 10,000 JSON values, bounded keys/strings/numeric tokens and a 16 KiB selected output cap. Content-Length is not used as a decoded-size oracle and there is no second decompression. These are local refusal limits, not completeness guarantees. Known original raw/URI/form key fragments are checked in public message fields and selected outputs; the intentional JSON key carrier is added separately. Earlier inputs and history are outside a universal-redaction claim.

The provider budget is ten seconds after input and credential preparation, covering public scans, request, response reading and projection. The shared gateway budget is eleven seconds after credential resolution; preparation has no deadline. Errors after a request is attempted are indeterminate, even when the block retry setting asks for more tries. Inspect Mailchimp Transactional before deliberately repeating a send. Stop cannot recall an email. Native local tests use synthetic transport; they do not establish real billing, delivery, matching-revision Review, organization authorization, durable execution or human consent.

LingvaNex Cloud v3

Save a native lingvaNexCloudV3Api account with exactly {"apiKey":"your-existing-key"} in Advanced JSON, then select that saved account. This explicit raw-key Cloud v3 profile differs from the shipped n8n Bearer header and has no automatic lingvaNexApi conversion or fallback. Settings checks original received JSON before outer trimming: local bounds are 32,768 JavaScript units and 1–4,096 visible non-space ASCII key characters. Those bounds do not identify a vendor key format. Save is local and unverified with no request; Test is unavailable. The lower generic nonempty store can retain malformed data that runtime rejects. Identity, privileges, key validity, entitlement and exact charges remain unknown.

Enable the LingvaNex workspace live control; an explicit workspace choice overrides WORKFLOW_LIVE_LINGVANEX. List languages requires initially unchecked acknowledgeProviderUsage:true and separate invocation confirmation. It may consume provider usage. The output is one bounded response of at most 256 unique language codes and English names, plus local coverage single-response. It does not certify a complete catalog. Translate text requires literal text of 1–8,192 JavaScript units, explicit from and to codes matching the local en_GB-style grammar, initially unchecked acknowledgeTextDisclosure and acknowledgeUsageCharges, and separate confirmation. Whitespace, Unicode and controls remain literal; HTML is treated as text. No auto-detection, HTML mode, batch/chunking, language lookup on Save or transliteration output is included. Whole expressions must preserve original string and boolean types; every acknowledgment must evaluate to actual true.

The fixed HTTPS Cloud v3 language and translation routes use the selected raw key, not Bearer. Translation sends fixed platform api and disables transliteration. Only actual HTTP 200, exact response URL, supported JSON media/content coding and own err:null plus the expected result shape succeed. Languages project only {code,name}; translation projects only {text}, including actual empty text. Unknown raw fields are parsed then discarded, with no invented source language, accuracy, timestamp or success flag. Returned text is not markup or instructions.

Local limits include 64 KiB request/output, 256 KiB exposed decoded response, 4,096 reads including completion, JSON depth 64 / 20,000 values and keys/32,768-unit raw or decoded tokens/262,144 scanner steps. Quoted-token limits include quotes and escapes, so not every nominal output maximum is accepted. Fatal UTF-8, duplicate decoded keys, malformed Unicode, BOM and trailing JSON are refused. No double decompression or decoded-size inference from Content-Length. Known raw/URI/form key fragments in public inputs and selected output cause refusal; short-key collisions may conservatively refuse otherwise valid data. Earlier inputs/history are outside a universal-redaction claim.

One attempt for either operation, even when Retry On Fail asks for more. Provider execution is bounded to ten seconds after local preparation; the gateway has eleven seconds after credential preparation, which itself has no deadline. A translation may process text or incur charges despite abort, Stop or a rejected response. An indeterminate attempt must not be repeated automatically. There is no hidden polling, readback, alternate host/authentication fallback or account probe. Local synthetic checks do not establish matching-revision hosted Review, organization/durable execution, human consent, real key rights or translation quality.

Magento 2 selected merchant token

Save native magento2Api with exactly {"host":"https://your-merchant.example","accessToken":"your-existing-token"}, then explicitly select that saved account. The local saved origin must be canonical HTTPS with a DNS hostname and no path, trailing slash, port, IP literal, user information, query or fragment. Origin syntax is not proof of public reachability; the server guard independently applies its network policy. Original JSON is capped at 32,768 JavaScript units, host at 2,048 units and visible non-space ASCII token at 1–2,048. These are local support limits, not a vendor format or permission guarantee. Save is local/unverified with no request and Test is unavailable; the shipped modules probe is not invoked. Generic Save may retain malformed nonempty data that runtime rejects.

This profile sends an existing Bearer access token to a selected Magento Open Source/Adobe Commerce PaaS merchant. It cannot identify token category, expiry, merchant privileges or whether standalone integration-token Bearer use is enabled. It does not acquire OAuth, mint/refresh tokens, alter merchant configuration or support Commerce SaaS. The explicit default-store route family is fixed; customer create/update retain their shipped no-store-code paths. No arbitrary URL/header/store, filters, loader, returned-URL following or automatic n8n conversion is available. An explicit workspace choice overrides WORKFLOW_LIVE_MAGENTO2.

Six reads cover customer/order/product get and one explicit list page. Lists require integer pageSize 1–100 and currentPage 1–1,000,000; no automatic pagination, complete export, stable snapshot or has-more claim. Every operation requires the initially unchecked acknowledgePersonalData:true because the raw body can contain personal data before projection, plus separate invocation confirmation. Nine effects additionally require acknowledgeSideEffects:true and their operation-specific account/email, overwrite, deletion, financial, invoicing/cancellation or fulfillment/inventory acknowledgments. All start unchecked and have no default. Callback true in local tests is not human consent.

Customer create/update require explicit email, first/last names and website ID; update also requires customer ID. They can affect accounts and email, and update does not promise to preserve omitted fields. Product create requires explicit SKU/name/attribute-set ID, original decimal price, simple type, status 1/2, visibility 1–4 and actual saveOptions:false. Product update is an explicit name-only save with saveOptions:false; both may create or update and may apply merchant financial defaults. Invoice/ship require 1–10 explicit unique {orderItemId,quantity} records and actual notify:false/appendComment:false; invoice also requires capture:false. These flags do not suppress unknown merchant hooks. No empty-lines full-order default is accepted.

Positive IDs are original canonical decimal strings bounded locally to9,223,372,036,854,775,807; counts can be0. SKU is an original single component of 1–64 ASCII letters/digits/dot/underscore/hyphen, starting with a letter or digit. Price 0–999999.999999 allows at most six fractional digits; nonzero quantity up to 1000 allows four. These original decimal strings become numeric JSON tokens without local floating-point conversion. That preserves the lexical request, not merchant storage width, float/currency rounding or monetary accuracy. Whole expressions retain actual booleans, decimal strings and item arrays.

Only actual HTTP 200 with exact selected URL and supported bounded JSON representation succeeds. Customer reads expose IDs only; order reads expose entity_id and optional actual string/null status; product reads expose ID/SKU. Lists retain those selected fields and optional actual total_count as an exact decimal string, including0. Correlated customer/product saves expose only selected IDs and explicit required echoes. Deletion/cancellation return actual JSON boolean, including false; invoice/shipment return only actual original numeric ID lexemes. None proves target state, payment, delivery, inventory change, existing-only save or success inferred from truthiness. Monetary and unknown personal fields are discarded; there is no hidden readback.

Local bounds are 64 KiB request,256 KiB exposed decoded response, 1,024 reads, JSON depth 16 / 5,000 nodes, 32,768-unit strings/65,536 raw string span / 256-unit keys/128-unit numeric lexemes and 16 KiB selected output. Fatal UTF-8, malformed Unicode, duplicate decoded keys and unsupported original numeric IDs refuse; valid U+FFFD string data remains literal. No raw/decoded Content-Length equality or second decompression. Known original raw/URI/form token fragments are scanned in public input keys/string values before controlled numeric carriers and in selected output; array indices/false/0 are not coerced into strings just to scan. Earlier unresolved-credential validation is outside universal secrecy guarantees.

All fifteen operations make one attempt even if Retry On Fail requests more. Provider execution has ten seconds after local preparation; gateway has eleven seconds after credential preparation and preparation has no deadline. An attempted effect can finish remotely despite timeout, Stop, malformed response or local persistence failure. Inspect the merchant before deliberately repeating an indeterminate write; no automatic retry or recall resolves uncertainty. This finite native profile is partial service support and local synthetic evidence, not real merchant, hosted Review, organization, durable execution or human-consent certification.

Home Assistant actions, Matrix, Mailcheck, Monica, Mindee and MessageBird

These six native profiles expose 47 finite operations. Save the exact account type, explicitly select that saved account on the block, and enable its workspace live-execution control. An explicit workspace value overrides the matching operator switch. Save sends no provider request and remains unverified; Test connection is unavailable. Settings validates the original received JSON before outer trimming. The lower generic nonempty store may retain malformed data that runtime later refuses. Local JSON, visible-ASCII secret and origin bounds are support limits, not vendor key formats or evidence of account identity, permissions, expiry or product availability. There is no OAuth issuance/renewal or automatic conversion of historical n8n credential profiles.

Applicable disclosure/effect acknowledgments start unchecked, remain editable, have no true default and must evaluate to actual boolean true. They remain separate from invocation confirmation and account selection. Whole expressions preserve actual scalar/object types; typing null as ordinary text creates a string. Use {{ null }} for an explicit nullable value. Required ordinary business booleans such as completed and delible must be stored explicitly, including false. A callback returning true in local tests is simulated confirmation, not human consent.

Home Assistant household actions and camera

The nine-operation actions profile is separate from the existing five Home Assistant reads and their account. Save a public HTTPS DNS origin and existing bearer token; private IPs, non-HTTPS origins and custom ports are unsupported, and the server guard independently checks egress. Its original credential JSON is bounded to 16,384 units/UTF-8 bytes, token to 16–8,192 locally supported characters, and selected context to 1–128 ASCII characters.

Actions are configuration-check, state-representation-set, event-fire, history-window, template-render, error-log, service-call, logbook-window and camera-image. Review household data, displayed-state replacement, event/automation, server-template and possible physical device effects. Setting displayed state does not prove a physical device changed. Service calls require explicit domain/service and a bounded raw bodyJson with entity_id; original JSON numbers remain original request tokens. History/logbook require one entity and an explicit UTC window of at most 24 hours. At most 250 history states or logbook/service entries are selected; error-log selects the last 1–100 lines. These outputs are bounded received data, not complete history or independent effect verification.

Camera reads one selected camera and stores an admitted PNG/JPEG in Workspace Files. It requires both camera disclosure and storage-effect policies plus separate confirmation. Output contains only fileRef/name/MIME/size; inspect Workspace Files before repeating an uncertain request or storage attempt. Local limits are 5 MiB, 4,096 reads, 8,192 pixels per dimension and 16 million pixels, with structural PNG/JPEG checks rather than pixel decoding. Files stay workspace scoped and can retain run references. Storage, download permissions, real scanner, retention and durable persistence depend on the deployment; a successful local memory-file test does not certify those services.

Matrix Client-Server v3

Use one canonical HTTPS homeserver root origin and existing access token. No custom path/port, IP origin or token acquisition is supplied. Nine operations cover account.me, event.get, messages.page, members.list, room.create/join/invite/kick/leave. All require the personal-data policy and separate confirmation; room and membership changes have their additional policies, and public directory listing requires its conditional acknowledgment. A public room preset and public directory visibility are different choices.

Room/event/user IDs and pagination cursors remain original strings. Empty optional cursors or reasons remain empty when explicitly entered. Lists return one bounded response; there is no automatic pagination or membership export. Encrypted events expose selected metadata only, without ciphertext decryption or HTML fallback. Actual false, timestamp zero and empty state keys remain distinct. Timestamp integers are checked against their original lexeme before numeric conversion. This profile accepts absent/identity content coding only. Server identity, device rights, membership privileges and current capabilities remain unverified; room creation or invitations can have lasting effects.

Mailcheck email and quota

One email-check sends exactly one locally supported ASCII email to the fixed legacy endpoint with Bearer authentication. Both personal-data and quota policies start unchecked and require separate confirmation. The response must contain the exact requested email and at least one supported signal. Only email, optional githubUsername, five optional boolean signals and original safe integer trustRate are projected; false and zero are retained. This is a partial legacy profile, not a current compatibility, deliverability or consent guarantee. It can consume paid quota; inspect an uncertain check before intentionally repeating it. Response limits are 32 KiB decoded, 128 reads, depth8/256 nodes and 8 KiB output; identity coding only.

Monica Classic4.x relationship records and reminders

Twenty-five fixed-origin operations cover tags get/page/create/update/delete, notes get/page/create/replace/delete, journal get/page/create/replace/delete, tasks get/page/create/replace/delete and reminders get/page/create/replace/delete. Save only the existing apiToken; deployed Classic version, account ownership, privileges, recovery and persistence remain unknown. Lists require an explicit page/limit and return one partial page without totals or completeness inference. IDs remain exact positive decimal strings of 1–15 digits; actual numeric response IDs are validated before conversion.

Read the relationship-data and note/journal/task/reminder disclosures, then check applicable change/deletion/overwrite policies and confirm independently. Tag/note/journal deletion also has the existing ordinary confirmation:true policy field. Task completed:false is explicit; completing can reset server completion time and clearing can remove it. Tasks create accepts explicit nullable contactId; replacement uses nullable contactIdFilter as a selection filter and neither reassigns nor unlinks the contact. Task description null, empty text and string 0 are distinct inputs, while PHP empty handling can yield actual server null; only the received description is returned.

Reminder replacement requires a NONNULL contactIdFilter selecting the current association; it never reassigns or unlinks it. Create/replace require explicit title, nullable description, frequencyType (one_time, week, month, year), frequencyNumber1–1000, Gregorian YYYY-MM-DD initialDate and ordinary delible:false or true. The positive interval bound and one_time number1 are local narrowing, not vendor defaults. Read recurrence accepts actual zero or null. Empty reminder description can become server null while string 0 remains distinct. Do not apply Tasks empty handling to reminders. A false delible flag is not proof of a deletion veto.

Reminder writes invoke source scheduling for EVERY account user, delete existing scheduled outboxes and can create reminder and advance-notification outboxes under active rules. Past/today dates roll forward, including year addition for one_time on that path. No extra recurrence/deletability acknowledgment replaces the existing reminder disclosure and independent confirmation. Requested past dates, delivery time/channel, per-user row counts, transactionality, timezone/month/leap rollover, server termination and cancellation of dispatched effects are not guaranteed. A client timeout does not cancel server computation. Selected received rows preserve actual nullable/empty description, zero/null recurrence and false delible; initial date, next-run, names and raw account data are omitted. Delete returns only a correlated actual acknowledgment, not proof of hard deletion or canceled delivery.

Mindee V1 Invoice V4 document processing

invoice-predict-v4 uploads one explicitly selected, owned Workspace File to the fixed V1 InvoiceV4 endpoint with Token authentication. Document-disclosure and paid-processing policies start unchecked and require separate confirmation. PDF/PNG/JPEG snapshots up to 1 MiB with admitted signatures are supported. The actual server file adapter may first buffer/hash an owned file up to its 100 MiB shared limit before this provider refuses an oversized snapshot; 1 MiB is an upload admission limit, not a pre-read memory cap.

Multipart has one document part with a generated filename and exact snapshot bytes; the original private filename is not sent. The runtime-generated multipart boundary is not inspected. Only HTTP200/201, actual api_request status success and the admitted document/invoice shape succeed. Missing status_code is allowed; when present it must be an original numeric200–299 integer. Selected invoice text, nullable values, document ID, original safe page count and independent locale/currency are returned. Money, line items and raw URLs are omitted. OCR accuracy, credits, document retention, product availability, accounting correctness and cancellation/refunds remain unverified. No ReceiptV5 or V2 fallback is included.

MessageBird Classic REST SMS and balance

sms-send sends one explicit originator/recipient/message with SMS type, automatic coding and mclass1. Local recipient support is a plus-prefixed 6–15 digit number and body up to1,600 units/6,400 UTF-8 bytes. SMS disclosure and paid-sending policies start unchecked and require separate confirmation. A message can become multiple separately billed parts. HTTP201 projects only the actual message ID, without delivery/recipient status or charges. There is no batch, voice, webhook callback or URL following.

balance-get requires its account-data policy and separate confirmation. It returns only actual payment mode, type and the original decimal amount as a string, preserving zero and trailing decimals without floating-point rounding. This is selected account data, not a financial balance verification or spending authorization. Both operations use fixed classic REST routes with AccessKey authentication. Any root errors member is refused locally, even if empty; this is conservative local support, not a universal vendor schema rule.

Shared limits and uncertain attempts

Each of the 47 operations makes one attempt, including reads, even when Retry On Fail requests more. Provider execution/read/projection budgets are ten seconds after confirmation and credential preparation; gateway budget is eleven seconds after credential resolution, and preparation itself has no deadline. Mindee file reading is inside its provider budget. Strict UTF-8, original JSON duplicate-key/depth/node/token/iteration limits and selected output bounds apply; unknown raw fields are discarded rather than dumped. There is no automatic pagination, hidden readback, alternate authentication or returned-URL follow. Accepted profiles keep their narrower content-coding rules; decoded response caps do not imply equality with raw Content-Length or a second decompression.

An attempted household, membership, CRM/reminder, quota, OCR or SMS effect may complete despite timeout, Stop or rejected local output. Inspect the service/Workspace Files before deliberately repeating an uncertain effect. Local synthetic tests do not establish matching-revision Review, organization/durable execution, human consent or real provider permissions, effects, delivery or billing. These remain partial native profiles, not all n8n operations or parameter conversion.

Receipt V5, Monica Calls and Mocean

These eight operations extend the preceding 47-operation group to 55 operations across eight native account profiles. Monica now has 30 operations. Receipt V5 is a separate Mindee account profile alongside Invoice V4; Mocean is a new provider. They remain finite subsets of the corresponding n8n integrations. No automatic n8n parameter or credential conversion is provided.

Select the exact saved account on the block. Receipt and Invoice accounts are offered only for their corresponding operations; neither profile substitutes for the other. Mindee's workspace control covers both profiles. An explicit workspace setting overrides its operator switch. Save sends no provider request and remains unverified. Test connection is unavailable. Settings checks the original received JSON before trimming outer whitespace, including the 32,768 JavaScript-unit limit. Local visible-ASCII secret bounds of 1–4,096 characters are support limits, not vendor key formats or evidence of identity, permissions or expiry. Mocean additionally bounds original JSON to 65,536 UTF-8 bytes. Generic nonempty storage can retain malformed material refused at execution. No OAuth acquisition or renewal is included.

Mindee Receipt V5

receipt-predict-v5 uploads one explicitly selected owned Workspace File to the fixed V1 expense_receipts V5 endpoint using Token authentication. It does not fall back to Invoice V4 or another product. The document-disclosure and paid-processing acknowledgments start unchecked and require actual true plus separate invocation confirmation.

The returned PDF/PNG/JPEG snapshot must fit 1 MiB and an admitted signature. The actual shared file adapter can first buffer and hash an owned file up to its 100 MiB limit; 1 MiB is the upload admission limit, not a pre-read memory cap. The upload copies the snapshot into one document multipart part with generated document.pdf/png/jpg name, without sending the private original filename. Signatures do not validate the complete document; the runtime-generated multipart boundary is not inspected.

Only HTTP200/201 with actual api_request.status:success and the admitted response shape succeed. Optional status_code must be an original numeric200–299 integer. Output selects documentId and optional documentName, nPages, receiptNumber, supplierName, time, currency and locale. Missing fields stay absent; admitted null, empty text and string 0 stay distinct. Page count uses an original positive safe-integer token before conversion. Locale and currency are independent, with no language fallback. Money, dates, line items, OCR text and raw URLs are omitted. OCR accuracy, processing credits, retention and refunds are unverified. Responses are bounded to 1 MiB decoded, 4,096 reads, depth16/10,000 nodes and 32 KiB output.

Monica Calls

calls.get, calls.page, calls.create, calls.replace and calls.delete use the existing Classic4.x account and fixed calls routes. Every action requires relationship-data and call-disclosure acknowledgments, initially unchecked, and independent confirmation. Create adds record-change acknowledgment; replace adds record-change and overwrite; delete adds record-deletion. There is no extra date/direction acknowledgment or new contact-selection policy.

IDs remain positive decimal strings of 1–15 digits. Get/replace/delete select callId. Page requires numeric page1–1,000,000 and limit1–25; it returns one partial response in received order, including an empty page, without totals or automatic pages. Create requires contactId; replace has no contact selector, contact filter, reassignment or unlinking. Create/replace require an exact Gregorian calledAt date from 0001-01-01 through 9999-12-31, required nullable content, and required nullable boolean contactCalled. Content supports empty text and string 0, up to8,192 JavaScript units/32,768 UTF-8 bytes. Store false explicitly; use {{ null }} for null through a whole expression, since ordinary text null is a string.

Received rows select only actual id, associated contactId, content and contactCalled. Server null/false/empty values are returned rather than echoed from the request. Replacement can apply PHP empty handling to empty text, string 0 or false; actual received values remain authoritative. Called dates, contact names, emotions and raw relationship data are omitted. Get/replace must correlate the requested call ID; create must correlate the selected contact. Delete returns an actual correlated acknowledgment, not proof of durable hard deletion.

The examined Classic source can update the associated contact's last_talked_to on create/replace using the later of the previous value and selected date; moving a call backwards does not recompute all history. Delete detaches the call's emotion association, removes the call and can set last_talked_to from the first remaining call or null. Remaining-call ordering is unverified. This does not establish deletion of global emotion definitions. Inspect the relationship record after an uncertain write. Response bounds remain 256 KiB decoded/4,096 reads/depth32/16,384 nodes; request and selected output each have a64 KiB bound.

Mocean V2 SMS and voice

sms.send and voice.say each send to one manually entered recipient. Both require the initially unchecked disclosure and paid-processing acknowledgments plus independent invocation confirmation. Use digits without a leading plus: recipient/caller numbers support6–15 digits and begin nonzero. SMS sender supports1–15 ASCII letters/digits/internal spaces and message1–1,600 units/6,400 UTF-8 bytes. Voice uses a caller number, message1–1,000 units/4,000 bytes and one explicit language: en-US, en-GB, cmn-CN, ja-JP or ko-KR. These are local support bounds, not delivery eligibility checks.

Requests are single fixed HTTPS form posts to /rest/2/sms or /rest/2/voice/dial. Both secrets are form fields; no arbitrary host/header/URL, batch, callback or returned-URL following is supplied. Voice sends one say command. SMS accepts HTTP202 with one status0 message for the exact recipient and selects messageId/recipient. Voice locally accepts HTTP200 or202 with exactly one status0 entry in the documented calls envelope and selects sessionId/callId; it does not reuse the shipped SMS envelope for voice. Accepted IDs do not prove delivery, a connected call, language quality or charges.

Responses are bounded to256 KiB decoded/4,096 reads/depth32/16,384 nodes; request/output each64 KiB. Original JSON duplicate keys and budgets are enforced. The Mocean parser uses objects without a prototype; it does not claim a reserved-name blacklist. Unknown fields are discarded after bounded parsing rather than exposed.

Attempts, privacy and remaining verification

Each of the eight actions makes one attempt, including reads, even when Retry On Fail requests several. Provider execution, file reading, response reading and projection have ten seconds after confirmation and credential preparation; the gateway has eleven seconds after credential resolution. Preparation itself has no deadline. A timeout, Stop or rejected result does not undo an attempted CRM change, document processing, SMS or call. Inspect the service before deliberately repeating an indeterminate effect.

Strict UTF-8 and original JSON depth/node/token/read limits apply before selected output. Authentication fragments are checked independently in actual object keys and string values using raw, URI and form representations from each raw secret; there is no chained encoding or primitive coercion. Fixed origins, exact returned Response URL, decoded-body caps and accepted content-coding rules remain enforced. Decoded caps do not imply raw Content-Length equality or a second decompression. Local synthetic tests do not certify matching-revision Review, real providers, organization/durable execution, billing or human consent.

Mistral OCR and Microsoft Teams reads

These nine actions extend the preceding 55-operation group to 64 finite operations across ten native account profiles. Mistral adds one PDF OCR action; Microsoft Teams adds eight Graph v1 reads. The ordinary and Tool reference nodes have partial service coverage only. Embeddings, chat models, Teams writes, triggers and send-and-wait are outside this group. Reconfigure supported fields explicitly; there are no automatic n8n import aliases or credential conversions.

Select the exact saved account on the block and enable its workspace control. An explicit workspace value overrides the operator switch. The Teams native profile accepts only microsoftTeamsGraphV1Api, not the historical microsoftTeamsOAuth2Api profile or a generic account name. The Mistral provider identifier is mistralAi; its normalized account family is mistralai. Neither identifier is a secret input field. Save sends zero provider requests and remains unverified; Test connection is unavailable. No token issuance, OAuth refresh, account discovery, scope probe or alternate cloud is included.

The existing workspace Teams OAuth connection remains separate. Its callback saves microsoftTeamsOAuth2Api credentials in the OAuth token format; it does not convert them into the native microsoftTeamsGraphV1Api account or make them selectable for these native operations. Create and select the manual native account explicitly when using this group.

Settings validates original received JSON before outer trimming, with a32,768 JavaScript-unit bound. Local visible-ASCII secret bounds are1–2,048 characters for Mistral and1–8,192 for Graph; Graph also caps original JSON at65,536 UTF-8 bytes. These are local support bounds, not vendor token formats. Generic nonempty storage can retain malformed material refused at execution. Identity, expiry, privileges and account eligibility remain unverified.

Mistral PDF OCR

document-extract-text requires one explicitly selected owned Workspace File ID, document-disclosure acknowledgment and paid-processing acknowledgment. Both checkboxes start unchecked, require actual true, and are independent of invocation confirmation. It sends one fixed JSON POST to https://api.mistral.ai/v1/ocr using Bearer authentication, the mistral-ocr-latest alias, pages0–7 and include_image_base64:false. The alias is not a pinned model version. There is no external-document URL, file upload endpoint, loader or returned-link fetch.

The returned owned snapshot must be PDF bytes with the admitted PDF prefix, an admitted PDF filename and MIME type, and size5bytes–1MiB. The actual shared file adapter can first read and hash up to100MiB; the1MiB limit applies to the returned snapshot before provider upload, not pre-read allocation. A foreign workspace file fails the scoped service lookup before blob opening. These checks do not validate a complete PDF. Bytes are copied and embedded as a base64 PDF data URL; the private original filename is not sent.

Only an actual HTTP200 response with the admitted shape succeeds. Selected output is exactly extractedText and pageCount. One through eight returned pages must have distinct ascending original indexes0–7 and string markdown; empty markdown remains empty. Text joins returned pages with two line feeds. pageCount counts the returned selected pages, not all document pages. Per-page text is capped at32,768 units; joined text at131,072 units/524,288 UTF-8 bytes. Raw usage, model metadata, image data and provider diagnostics are omitted. OCR correctness, retention, charges and refunds remain unverified. Treat extracted text as data rather than rendered Markdown or instructions.

Request and exposed decoded response each have a2MiB bound. The original-JSON reader enforces strict UTF-8, duplicate keys,4,096 reads, depth32,20,000 nodes, bounded arrays/objects/keys/string and number tokens before projection. Mistral accepts the shared bounded known content-coding rules; it does not claim identity-only encoding, raw Content-Length equality or a second decompression.

Microsoft Teams Graph v1 reads

All eight actions require the initially unchecked data-access acknowledgment and independent invocation confirmation. Use manually selected opaque IDs as exact strings,1–512 visible-ASCII characters, excluding slash, backslash, query/fragment separators and dot segments. No ID, account or resource discovery request is made.

The fixed HTTPS origin is https://graph.microsoft.com. Channels use a fixed $select of id,displayName,description,membershipType. Message pages require an explicit numeric pageSize1–25 and fixed $top. Planner pages do not invent $top or perform a hidden identity lookup. Channel-message reads use the documented v1 route rather than the shipped beta route.

For a first page, continuation is required null: use {{ null }} in the ordinary text editor, because literal text null is a string. Later supply the complete supported nextPageUrl explicitly. The full URL must preserve the same manually selected resource route and fixed query, with exactly one supported $skiptoken or $skip marker and bounded unique keys. Only literal colon/@ equivalence to their encoded spellings is allowed within opaque IDs. The received URL is sent unchanged. A missing returned next link becomes null; an explicit null next link in the provider envelope is refused. There is no automatic follow, retry, completeness guarantee or arbitrary URL/header input.

Only an actual HTTP200 succeeds. Get returns a selected object; page returns items plus nextPageUrl, including an empty items array. Channel output selects id and admitted displayName/description/membershipType. Message output selects id, replyToId, subject, three admitted timestamps and body.contentType/content. HTML stays opaque text; attachments, images, senders, threads and links are neither fetched nor rendered. Planner output selects id/title/planId/bucketId/etag and optional percentComplete. That percentage must be an original canonical numeric integer0–100; false, null, quoted numbers, fractions, exponents and negative zero are refused. String IDs retain their original spelling, including values larger than safe JavaScript integers. Admitted null, empty text and string0 remain distinct from missing fields.

Responses require identity-only content coding, at most1MiB exposed decoded bytes,4,096 reads, depth32,16,384 nodes, bounded keys/tokens and65,536-byte selected output. Rights differ among channels, channel messages, chats and Planner. Work/school eligibility, membership, consent, licensing, scopes, expiry and availability remain unverified. Personal accounts and national-cloud origins are unsupported. This group adds no sends, writes, callbacks, human-in-the-loop lifecycle or loaders.

Attempts and remaining verification

Every action makes one attempt, including reads and Retry On Fail configurations. Provider execution, owned-file reading, response reading and projection have ten seconds after confirmation and credential preparation; the gateway has eleven seconds after preparation. Preparation itself has no deadline. Stop or timeout cannot undo attempted OCR processing or charges. Inspect an indeterminate OCR effect before deliberately repeating it.

Original JSON budgets apply before selected projection. Secret checks inspect actual object keys and string values, using independent raw, URI and form fragments from each raw secret with duplicates retained; arrays contribute values, not indexes, and primitives are not coerced. Fixed hosts, real Response objects and exact returned Response URLs remain required. Static errors omit unknown input keys, response bodies and provider diagnostics. Local synthetic tests do not certify providers, human consent, organization scope, durable execution or a matching-revision Review walkthrough. User-authorized hosted access is distinct from deploying and verifying this dirty revision; matching-revision Review and cohort provider credentials remain pending.

Monica Activities and Teams mutations

Fourteen further actions extend the preceding64-operation group to78 finite operations across the same ten native profiles. Monica now has35 actions and global Microsoft Teams Graph v1 has17, including its eight existing reads and nine new mutations. These are partial ordinary/Tool service subsets, with no automatic n8n import alias or credential conversion. The earlier nine-action Mistral/Teams-read section describes that read subset; the mutation subset below is now separately available.

Keep Monica's selected monicaClassic4xApi account with exact apiToken, or Teams' selected microsoftTeamsGraphV1Api account with exact accessToken. Existing workspace controls remain WORKFLOW_LIVE_MONICA_CLASSIC4X and WORKFLOW_LIVE_MICROSOFT_TEAMS_GRAPH_V1. Save is local/unverified with zero provider HTTP; Test is unavailable. No OAuth acquisition/refresh, hidden identity lookup, ETag lookup or prerequisite read is added. Rights, deployed version, account eligibility, membership and token expiry remain unverified.

Monica Activities5

Every action requires initially unchecked acknowledgeRelationshipProcessing and acknowledgeActivityDisclosure, both actual true, plus independent confirmation. Create/replace add acknowledgeRecordChange; replace also requires acknowledgeOverwrite; delete requires acknowledgeRecordDeletion. Activity and participant data concerns private relationships and may be logged.

The fixed origin is https://app.monicahq.com. Manual IDs are canonical positive decimal strings of1–15 digits, not numbers. activityTypeId is required nullable; contactIdsCsv is1–10 unique IDs separated by ASCII commas, with no spaces or empty set. happenedAt is an exact Gregorian YYYY-MM-DD from0001 through9999, sent without timezone conversion. summary is nonblank text up to255 JavaScript units/1020 UTF-8 bytes. description is required nullable text up to8192 units/32768 bytes: empty text, string0 and null remain distinct. Use {{ null }} for actual null, not ordinary text null. Replacement sends the whole participant set and all five fields in the fixed JSON shape, detaching omitted contacts.

Source creates a journal association before attaching participants. Replacement updates scalars, deletes ONE resolved associated journal, creates an association and syncs the whole attendee set. Participant sync rebuilds yearly statistics for DETACHED, ATTACHED and UPDATED contacts. Deletion removes ONE resolved journal before deleting the activity; it does not establish all-history purging, physical deletion or recovery. Transactionality, observers, broader ORM effects, cascades and persistence remain unverified. Empty descriptions can become null in the shown service; string0 is preserved.

Received activity output selects exactly id, summary, description and contactIds, preserving selected empty/null values and received contact order. Numeric IDs are checked using original integer lexemes before projection to strings. Pages return only items from one bounded response; totals, dates, types and other raw PII are discarded. There is no automatic pagination or completeness guarantee. Get/replace require matching activity IDs; create/replace conservatively require the returned participant SET to match the requested set. Stale, filtered or cached results can therefore be indeterminate after invocation.

The local delete acknowledgment requires an actual canonical STRING ID matching the requested activity. A numeric acknowledgment is refused after invocation with an uncertain outcome; actual remote wire type remains unverified. A selected deleted:true acknowledgment is not independent proof of durable deletion.

Monica's original received JSON is bounded to256KiB/4096 reads/depth32/16384 nodes, with bounded tokens, keys, numbers, parser work and64KiB request/output. Identity-only content coding is required. Secret checks cover the defined input, request and selected public output; this profile does not claim scanning every dropped private raw-response string.

Teams WRITE9

All mutations require initially unchecked acknowledgeDataAccess and independent confirmation. Create/send adds acknowledgeRecordChange; update also requires acknowledgeOverwrite; delete requires acknowledgeRecordDeletion. Message content may reach recipients. No workflow link is appended, and HTML remains opaque text locally.

Every route uses https://graph.microsoft.com/v1.0 with each manual opaque ID encoded once. No arbitrary URL/header input is accepted. Channel displayName is nonblank1–50 units; description permits empty text up to1024 units. membershipType is standard/private. ownerUserId is required null for standard, or an explicit UUID-form user ID for private; only the precise owner member shape is sent. Shared/async channels are excluded. A private-channel201 does not prove SharePoint provisioning.

Message contentType is exactly text/html; content is bounded nonblank1–8192-unit text, with no attachment, mention, link append or HTML rendering. Planner title is bounded nonblank1–255 units. percentComplete must be an actual canonical numeric integer0–100; preserve numeric0, not string0, false or null. Required nullable update values use {{ null }} when omitted; numeric0 remains a change. Channel empty description is an explicit change. Planner update/delete send the exact manually supplied weak/strong quoted ETag as If-Match; wildcard, hidden GET, refresh, fallback and automatic retry are unsupported.

Only the stated status succeeds. HTTP201 projects the actual received selected channel, message or task data through the existing read projection. It does not echo requested values as remote truth. HTTP204 requires an actual null response body and absent or exact zero content-length, returning exactly httpStatus:204; it does not prove resource state, delivery or durable deletion. Teams scans the whole bounded parsed response keys/string values before selected projection. Request/output are capped at64KiB; response is identity-only, up to1MiB/4096 reads with the accepted depth/node/token limits.

Attempts, consent and verification

Each of the fourteen actions makes one attempt despite Retry On Fail. Provider10s/gateway11s begin after confirmation and credential preparation; preparation itself has no deadline. Any failure after mutation invocation is uncertain. Stop/timeout cannot guarantee server cancellation or rollback; inspect remote state before deliberately deciding to repeat. No transaction, remote exactly-once, full pagination or full n8n parity is claimed.

The exact selected native credential and independent confirmation remain separate from initially unchecked field acknowledgments. Local synthetic tests and callbacktrue are not human consent, vendor permission, organization-scope execution, durable storage or matching-revision Review evidence. User-authorized hosted access exists, but deployment of this dirty revision and provider credentials remain pending. Existing Mistral limits and all earlier finite operation boundaries remain applicable.

Perplexity Agent and OpenThesaurus

These two finite actions extend the preceding78-action group to80. Perplexity adds one saved native credential profile; OpenThesaurus is public and has no credential, account, Save or Test placeholder. Both are partial ordinary/Tool service subsets with zero automatic n8n import aliases. Imported Perplexity Sonar chat-completions nodes are not converted into Agent runs.

Perplexity Agent fast

Choose Perplexity Agent → Ask Agent (fast), operation agent-answer. Save one existing API key as a native perplexityApi account and select that exact account on the block. Structured fields save JSON with exactly apiKey. Saving is local and unverified, sends no provider HTTP and does not infer account identity, plan, privileges, expiry or billing. Test connection, OAuth acquisition/refresh and a key-eligibility probe are unavailable. Generic nonempty persistence may retain malformed material; execution always rejects it before the shared gateway receives a credential.

Runtime validates the original received JSON at32768 JavaScript units, the key at1–2048 visible ASCII units and the separately selected context ID at1–256 units. These are local support limits, not vendor key-format rules. The normalized credential has exactly id/accessToken/sensitiveValues/metadata, with fixed provider/origin/auth metadata and raw/URI/form secret fragments independently derived from the raw key. A generic HTTP, OpenAI or OAuth account cannot substitute. Keep the key out of the prompt and expressions.

Enable the workspace Perplexity live control (WORKFLOW_LIVE_PERPLEXITY). Required input is prompt plus actual boolean true for initially unchecked acknowledgePersonalData and acknowledgePaidRequest; neither is independent invocation confirmation. Prompt preserves original nonblank valid Unicode, excludes NUL and is bounded to8192 JavaScript units/16384 UTF-8 bytes. Typed expressions preserve actual values and never turn a string true into consent.

Exactly one POST goes to https://api.perplexity.ai/v1/agent with the original Bearer key and fixed JSON preset fast, input prompt, stream false and max_output_tokens1024. This is explicitly distinct from shipped n8n's earlier Sonar operation, following the official Agent migration material. Backend/preset behavior can change. One HTTP request can cause multiple provider research/model/tool steps and API charges; it does not bound total spend or tool use. There is no model, endpoint, arbitrary header or URL selection, hidden probe, retry or source-URL fetch.

An actual unredirected200 Response with exact final URL, supported JSON/UTF-8 and bounded known content coding is required. The exposed decoded body is capped at2MiB/4096 reads. Content-Length is not compared with decoded length or claimed as raw wire-size proof. Original JSON refuses duplicate keys, malformed Unicode, depth above32, over20000 counted nodes and bounded token/key/numeric/work violations; numbers retain original lexemes and numeric IDs/usage are not projected. A bounded8MiB aggregate known-fragment ledger scans actual keys/string values, including ignored response subtrees, without array-index synthesis, coercion or recursive encoding. Unsupported responses and errors use static collision-aware messages.

Output is exactly answer and sources. Supported completed assistant output_text parts are joined locally with newlines in received order; this is a local aggregation, not a certified final-answer field. Root error/errors/refusal presence is rejected even null. At most32 selected title/url sources remain in provider order with duplicates retained. URLs must have original HTTP(S) spelling and ASCII LDH hosts; userinfo, explicit ports, Unicode hostname conversion and unsupported numeric/IPv6 hosts are refused. Sources are unnumbered and never fetched. Answer citation markers remain unverified model text without positional mapping to that list. Selected output is capped at256KiB.

OpenThesaurus German synonyms

Choose OpenThesaurus → Look up German synonyms, operation synonyms.lookup, and enable its workspace live control (WORKFLOW_LIVE_OPENTHESAURUS). Do not attach any saved account or configured credential reference. Only this exact provider/operation/tool, authnone and zero effective credentials may use the new public path, with all credential binding/reference carriers unset. Existing Hacker News public routes are preserved; this is not a general accountless-integration exception.

Input is word, actual boolean baseform, and actual true acknowledgeQueryDisclosure, initially unchecked, followed by separate confirmation. The word is nonblank valid Unicode with 1–128 UTF-16 units and up to 512 UTF-8 bytes, excluding C0/DEL. False baseform remains false. Each invocation creates a fresh query and makes exactly one GET to https://www.openthesaurus.de/synonyme/search with q, format application/json and explicit baseform true/false, honest contact User-Agent and identity encoding. No body, authentication, cookie, redirect, retry, pagination or follow-up URL request is sent.

Only an actual unredirected200 Response with exactly that requested URL is admitted. The exposed body is identity-only, at most512KiB/4096 reads; a declared canonical Content-Length must match those identity bytes. Fatal UTF-8, BOM, duplicate keys, malformed JSON/Unicode, depth above24, over20000 counted nodes and finite key/token/numeric/work violations are refused. IDs must be original canonical positive numeric integer lexemes1–15 digits and are projected directly to strings without Number rounding. At most32 synsets,128 terms per group and2048 total terms are selected, preserving order/duplicates and selected empty category/level strings. Error/errors presence on each selected record is conservatively refused, even null; bounded unknown fields are ignored. This does not claim all German meanings, ranking, locale detection, freshness or completeness.

Output is exactly synsets and attribution. Source must equal https://www.openthesaurus.de and actual nonblank copyright/license are required. The node output and inspector display the actual copyright/license and a visible OpenThesaurus source link without nofollow, across JSON/Table/Schema views. Provider strings are escaped text, never HTML, and opening the link is a separate user action. Empty synsets still retain attribution. A URL inside JSON alone is not the attribution display.

The public API terms request operator contact before ongoing use and publish an IP rate limit. This integration does not send operator email or implement a distributed quota limiter. The earlier accepted single public Auto example is captured evidence only, never a new provider-current or native-fetch guarantee.

Confirmation and deadlines

Each action makes one attempt despite Retry On Fail. Provider10s/gateway11s begin after credential preparation and independent confirmation; preparation/confirmation has no local deadline. Local abort/timeout does not prove remote cancellation, avoided billing or deletion of submitted data. A privacy-safe error message can be empty; execution still enters the failure branch, preserving configured on-error policies, and must not mistake it for successful output or a parked wait. Actual native fetch requires the fixed host/origin and received Response URL; arbitrary headers/URLs and returned-source following are unsupported.

A manually confirmed candidate run signs run/workspace/actor/workflow/version/material and exact operation binding, persists that evidence, and rechecks current explicit authority and live control. Perplexity additionally binds the selected account revision. OpenThesaurus uses a literal no-auth binding without credential fields; configured credentials, invalid signatures, changed material or replayed scope do not create a broader public grant. Organization membership alone does not authorize execution. Field acknowledgments, a saved account and workspace enablement do not replace separate confirmation.

Local synthetic tests cover ordinary controls, typed expressions, two workspaces, saved accounts, signed persisted grants, organization-context entry, refusal cases, actual server guard with synthetic responses and visible attribution. A true test callback is not human consent. Matching-revision Review, actual Prisma organization membership/durable execution, real Perplexity key eligibility/schema/spend/cancellation and native provider transport remain unverified. Hosted access is authorized; the missing matching deployment/provider material remains the specific blocker. No full n8n parity or automatic migration is claimed.

Outlook token reads and Monica conversations and contacts

These23 further actions extend the preceding80-action group to103 finite actions. Monica now has46 actions. Outlook adds a separate Microsoft Outlook Graph v1 token profile and twelve reads alongside the earlier OAuth Outlook subset. These ordinary/Tool service subsets add no automatic n8n import aliases or credential conversion. Sixteen further Outlook writes are described below.

Accounts and confirmation

Save one existing token as microsoftOutlookGraphV1Api with exactly accessToken, or reuse Monica monicaClassic4xApi with exactly apiToken. Select the saved account on the block. Outlook requires the exact native type; generic HTTP and historical outlookOAuth2Api cannot substitute. Existing Monica credential type normalization remains supported. Save does not contact the provider and leaves authentication, identity, permission and expiry unverified. Test and OAuth acquisition/refresh are unavailable. Generic Outlook JSON Save can retain a wrong-shaped object; strict execution refuses it. Structured fields serialize JSON. Advanced JSON retains typed text until formatting, switching to fields or server validation parses/reserializes it: duplicate keys can collapse to the last value and outer whitespace is lost, while extra keys remain. Strict runtime validation covers the received saved JSON, not an earlier envelope that was normalized away. Saving does not grant mailbox or relationship-data authority.

Execution validates original received JSON up to32768 JavaScript units/65536 UTF-8 bytes, a visible-ASCII token (Outlook1–8192; Monica1–4096) and the separately selected context ID1–256. These are local support bounds, not vendor token-format rules. Normalization yields exactly id, accessToken or secret, sensitiveValues and finite provider/origin/auth metadata. Raw/URI/form fragments derive independently from each raw secret, retaining duplicates. Do not put tokens or selected-context IDs in operation inputs.

Enable the distinct workspace control WORKFLOW_LIVE_MICROSOFT_OUTLOOK_GRAPH_V1 or existing WORKFLOW_LIVE_MONICA_CLASSIC4X. Explicit workspace false overrides an enabled environment switch. Applicable acknowledgment switches start unchecked and require actual boolean true, separately from invocation confirmation. Typed whole expressions preserve null, false, numeric0 and empty strings; text null is not null. Use {{ null }} for nullable values. Neither a true callback in local tests nor an acknowledgment proves human consent.

A confirmed manual run signs and persists exact run/workspace/actor/workflow/version/material/operation/tool/account-revision bindings. Changed account material, scope, signature or current-live false is refused. The exact23 new operations also recheck current-live state after callback confirmation and credential preparation, immediately before dispatch. Organization membership alone does not grant execution.

Twelve Outlook reads

Every action requires manually supplied nonzero lowercase mailboxUserId UUID and true acknowledgeMailboxAccess. Opaque resource IDs are1–2048 characters from the supported original ASCII alphabet; no hidden ID lookup or account-identity inference occurs.

Pages additionally require numeric pageSize1–25 and continuationUrl. Explicit null starts one page. A subsequent invocation may use a supported original next-page cursor with the same exact route, fixed selection and page size, and one skip/skiptoken parameter. Original cursor spelling, escapes, plus signs and padding are retained; this is not proof of cursor provenance. There is no automatic paging, full-mailbox/history completeness, arbitrary URL/header option, redirect or returned-link follow.

Exactly one GET goes to https://graph.microsoft.com using the selected Bearer token, fixed $select and identity encoding. An actual unredirected200 Response with exact nonempty final URL is required. Output contains only explicitly selected calendar/contact/event/folder/message scalars and nested owner/email/recipient/time fields, preserving supported null/missing/false/0/empty values, order and duplicates. Folder counts use original canonical bounded integer lexemes. Pages return their selected items and supported nextPageUrl; no raw mailbox response, attachments or HTML rendering is exposed.

Own error/errors keys on a selected root, row, email, recipient or time record reject the whole read even when null. Other unknown fields are scanned for known secret fragments then dropped. Fatal UTF-8, duplicate JSON keys, malformed Unicode and unsupported schemas/codings are refused. The exposed body is capped at1MiB/8192 reads, depth32,32768 nodes and bounded keys/tokens/numeric/parser work; selected output is capped at256KiB. This does not certify provider schema, mailbox rights, token eligibility, freshness or complete data.

Monica conversations6

All six actions require true acknowledgeRelationshipDataProcessing. IDs are original positive decimal strings of1–15 digits; dates are real Gregorian UTC timestamps in YYYY-MM-DDTHH:mm:ssZ form, without offsets or normalization.

• conversations.get: GET /api/conversations/{conversationId}.
• conversations.create: POST /api/conversations with contactId, contactFieldTypeId and happenedAt; also acknowledgeRecordChange.
• conversations.update: PUT the selected conversation with contactFieldTypeId and happenedAt; also acknowledgeRecordChange and acknowledgeOverwrite.
• conversations.delete: DELETE the selected conversation; also acknowledgeRecordDeletion. Associated messages may be deleted.
• conversationMessages.add: POST the selected conversation's messages collection with content, writtenAt and actual boolean writtenByMe; also acknowledgeRecordChange.
• conversationMessages.replace: PUT the selected message with the same three fields; also messageId, acknowledgeRecordChange and acknowledgeOverwrite.

Content is original nonblank valid Unicode1–8192 units/32768 UTF-8 bytes without NUL. False writtenByMe remains false. One request goes to fixed https://app.monicahq.com; no prerequisite read is added. Selected conversation output preserves server dates as received and up to128 complete ordered messages with duplicates. Replace requires the selected message exactly once with matching content/boolean, without assuming server date equivalence. Add does not invent a created message ID. Delete requires a200 deleted true acknowledgment with the original numeric conversation ID; it does not prove durable deletion or recovery.

Monica contacts5

All five actions require true acknowledgeRelationshipDataProcessing. Contact IDs are positive canonical decimal strings up to signed int64 maximum9223372036854775807, retained from original numeric lexemes without Number rounding.

Create POSTs /api/contacts; get/replace/delete use GET/PUT/DELETE /api/contacts/{contactId}. List GETs one explicit page with numeric page1–1000000, limit1–100 and sort created_at, -created_at, updated_at or -updated_at. There is no automatic pagination or complete contact-list guarantee.

Create and replace require all nine fields: firstName, lastName, nickname, description, genderId, isPartial, isDeceased, birthDate and deceasedDate. First name is nonblank; the nullable text fields preserve null versus empty strings. Gender is a nullable positive bounded integer string. Dates are nullable real Gregorian calendar days YYYY-MM-DD; a nonnull death date requires isDeceased true. False booleans remain false. Both require acknowledgeRecordChange. Replace additionally requires acknowledgeOverwrite and acknowledgeDateReminderReplacement: it clears/rebuilds selected birthday/death-date and reminder state, with newly requested reminders disabled.

Contact get can update consultation/view state and is disclosed as an external write even though its HTTP method is GET. Delete requires acknowledgeRecordChange and can affect relationships, orphaned partial contacts, avatar/soft-delete/jobs; completion and recovery remain unverified. A deletion acknowledgment requires200 deleted true with an original string contact ID, conservatively different from the conversation numeric-ID acknowledgment. Unsupported acknowledgment types refuse the result and leave an invoked mutation uncertain.

Limits and verification boundary

Each of these23 operations makes one attempt, including when Retry On Fail requests three. Provider10s/gateway11s start after independent confirmation and credential preparation, without a preparation deadline. Local abort/timeout cannot prove remote cancellation, rollback, deletion or avoided effects. Any invoked mutation with an unsupported/error outcome is uncertain; inspect remote state before deliberately repeating.

Monica uses actual exact-URL200 Responses, identity-only exposed bodies capped at256KiB/4096 reads, original JSON duplicate-key/Unicode/depth32/16384-node/token/work limits and64KiB request/output limits. Defined actual keys/string values are scanned within a finite budget, without coercion, array-index synthesis or chained encoding. This does not claim a scan of every dropped private Monica response string. Static collision-aware errors do not echo input keys, provider bodies or diagnostic messages.

Local synthetic tests exercise actual native gateway/runtime, two workspaces, distinct saved A/B tokens, signed persisted manual grants and mounted controls. Local UI Save responses are simulated; authenticated successful settings Save, real MFA/session/Prisma membership, durable organization execution, human consent, matching-revision Review and real provider requests/schemas/rights remain untested. Hosted access is authorized; these tests do not supply the matching deployment and provider material. No full n8n parity or publish-readiness claim is made.

Outlook token writes

Sixteen writes extend the preceding103-action group to119. The same Microsoft Outlook Graph v1 token profile now has28 finite operations: twelve reads and sixteen writes. Monica remains46. Use the exact saved microsoftOutlookGraphV1Api account and existing WORKFLOW_LIVE_MICROSOFT_OUTLOOK_GRAPH_V1 control. Save remains local and unverified with zero provider HTTP, and Test is unsupported. No OAuth acquisition/refresh, credential substitution or automatic n8n import conversion is added.

Every operation requires a manually entered nonzero lowercase mailboxUserId UUID and actual true acknowledgeMailboxAccess. Resource IDs use the original supported1–2048-character ASCII alphabet and are encoded once. No account identity or resource discovery is inferred. All applicable acknowledgment switches start unchecked; they require actual boolean true and remain separate from execution confirmation or an exact signed grant. A true programmatic callback is not human consent.

Required nullable selections must be supplied: use {{ null }} for typed null, not the text null. In draft updates, null omits the field and an empty string clears it. Draft subject/bodyText/toName explicitly preserve intentional empty or whitespace content; fields without that schema opt-in keep their existing normalization. Subject and recipient name reject control characters; the plain-text body additionally permits tab/newline controls. Subject supports up to255 JavaScript units, plain-text body up to16384 and recipient name up to128; these and the supported address/UTC grammars are local limits, not a claim of full vendor format support. Unknown input fields are refused.

One fixed request goes to https://graph.microsoft.com/v1.0/users/{mailboxUserId} beneath the selected calendars, contacts, mailFolders, calendar events or messages resource. No hidden GET, ETag, read-before-write, batch, arbitrary URL/header option, returned-link follow or automatic retry is added. Draft send POSTs only the selected existing message's /send route with no body;202 with no response body returns only httpStatus202. This observes acceptance, not delivery, completion in Sent Items or remote draft eligibility. Message update does not first prove isDraft. Create/update return only explicitly selected bounded scalars, preserving missing/null/false/empty; they do not dump raw mailbox data.

Deletion requires204. Calendar deletion permits an absent or bounded empty body; contact, folder, event and message deletion require an actual null body. Calendar/folder/event/message deletes return only httpStatus204. Contact deletion returns the selected request's mailbox/contact IDs and local deleteAcknowledged true; none proves durable deletion, retention or recovery. Selected event nested records reject own error/errors keys even when null. Unsupported statuses, URLs, schemas, known secret fragments or malformed/duplicate original JSON refuse the result. The received exposed body remains bounded at1MiB/8192 reads and selected serialized output at262144 UTF-8 bytes; local parser/projection bounds do not certify a provider schema.

All sixteen writes make one attempt, including at Retry On Fail3. Provider10s/gateway11s start after independent confirmation and credential preparation; preparation has no deadline. Cancellation or timeout cannot prove rollback or remote cancellation. An uncertain result after invocation can mean the write happened: inspect remote state before intentionally repeating. Current-live false and selected-account revision changes are rechecked after preparation before dispatch.

Local synthetic tests exercise the actual app/native gateway/runtime, saved accounts, two workspaces, signed persisted grants, mounted controls, finite deadlines and local cancellation. They do not prove authenticated successful settings Save, real MFA/session/Prisma organization persistence, human consent, matching-revision Review, live Graph rights/schema/notifications/retention/delivery or actual wire framing. Current draft-create documentation lists Content-Length0 alongside a required JSON body; that contradiction and automatic transport framing remain unresolved. No full n8n parity is claimed.

urlscan private scans and Marketstack v2 reads

Five actions extend the preceding119-action group to124. These are finite ordinary/Tool service subsets; no automatic n8n import alias, credential conversion, trigger or full provider parity is added.

Save and select the native account

Save an existing key using exactly urlscanV1Api or marketstackV2Api, each with JSON containing only apiKey. Named password fields serialize that object. For these two profiles, malformed original JSON is refused before trimming, formatting or switching to fields; duplicate keys cannot be repaired silently by those controls. Settings Save makes zero provider requests and leaves authentication, identity, expiry, rights and plan eligibility unverified; Test and OAuth acquisition/refresh are unavailable. Generic nonempty storage can still retain material that execution refuses. The exact native type and selected saved account are required; keep credentials and selected context IDs out of operation inputs.

Local support bounds are not vendor key-format guarantees: urlscan accepts16–256 ASCII letters, digits, underscore or hyphen in an original JSON envelope up to1024 UTF-8 bytes; Marketstack accepts1–2048 visible-ASCII key characters in up to32768 JavaScript units/65536 UTF-8 bytes. The selected account context is bounded separately. A prepared credential retains exactly id, accessToken, sensitiveValues and finite provider/origin/auth metadata. Secret fragments are independently raw/URI/form encoded with duplicates retained.

urlscan v1

Enable the workspace urlscan live control (WORKFLOW_LIVE_URLSCAN_V1 when no explicit setting exists). Submission discloses the URL to urlscan, asks it to visit and store the site, may consume quota, and lets the visited site observe the scan. Private visibility is requested, and anyone given a scan ID may gain access to scan data. The narrow HTTPS URL grammar does not establish public DNS or absence of personal data.

The submission disclosure starts unchecked and also requires independent execution confirmation. A failed submission after transport invocation has an uncertain outcome; inspect urlscan before deliberately repeating. There is no polling, automatic replay, hidden ID discovery or completion guarantee. The gateway allows10 seconds after credential preparation; this profile adds no separate provider timer or preparation deadline. Submission responses are capped at64KiB, result responses at1MiB and public output at4KiB, with finite UTF-8/original-JSON/depth/token/read/work limits. Unexpected status, envelope, correlation or exposed-content encoding is refused with static errors.

Marketstack v2

Enable the workspace Marketstack control (WORKFLOW_LIVE_MARKETSTACK_V2 when no explicit setting exists). Every query discloses the selected ticker/exchange and key and can use quota. Each acknowledgeQueryAndQuotaUse switch starts unchecked and requires actual true plus independent execution confirmation. The key is carried internally in the fixed HTTPS query; do not copy authenticated request URLs into workflow output.

All five actions make one transport attempt even when the node has maxTries3. There are no redirects, arbitrary URLs/headers, returned-link follows or automatic pages. Marketstack uses provider10s/gateway11s after confirmation and preparation, without a preparation deadline. Exposed decoded response bodies are bounded to1MiB with original JSON, UTF-8, key/token/depth/node/read/work and output budgets; supported declared content codings are checked without a second decompression or equating raw Content-Length with decoded length. Explicit selected zero/false/null/missing values remain distinct where supported. Unknown fields are dropped and static errors do not echo provider diagnostics or input keys.

Late changes and remaining verification

Both profiles check the current saved account and current live setting after both dispatch callbacks, using an immutable private stored-record fingerprint bound to the exact prepared object. In-place secret/name/type/owner/scope/workspace changes, missing accounts, copied prepared objects and current actor/live refusal stop execution. The fingerprint check itself does not resolve or decrypt the secret. This is an observation before dispatch, not an atomic transaction with the remote request or a preparation deadline. Stop/timeout does not establish remote cancellation, avoided quota or rollback.

With database persistence, failed credential, permission, run or live-setting reads stop these requests even when memory fallback is enabled elsewhere. Required current preference and run records must exist; cached account material cannot replace a failed database read. Memory-only local runs cannot verify persisted cancellation and still rely on the current runtime identity and Stop signal.

Local app/router/runtime and mounted-control checks use synthetic Responses and simulated consent/Save. They do not verify a live provider, current vendor wire compatibility, successful authenticated MFA Save, Prisma organization execution, human consent or a matching-revision Review walkthrough. Marketstack pre-provider confirmation/context callback privacy gaps in the bounded accepted candidate remain unresolved; no blanket secrecy guarantee is made.

npm public registry metadata

The npm public registry block provides four bounded reads without an account, credential, Save/Test connection or authentication fallback. Each read requires independent execution confirmation; no acknowledgment input is added. WORKFLOW_LIVE_NPM_PUBLIC_V1 and the workspace live setting control execution. The ordinary npm and npm Tool services have partial coverage with zero automatic import aliases; historical npmApi credentials and imported shapes are not converted.

Names are lowercase ASCII, 1–214 characters, optionally @scope/name; the entire name is encoded as one path component. Versions are explicit three-part decimal versions with bounded prerelease/build syntax, at most 128 characters; ranges, v prefixes and latest are refused. Search accepts 1–128 ASCII characters in up to eight simple words, with one space between words; operators are unavailable. Fields are exact and typed, with no coerced defaults, extra input keys or account bindings. Optional selected empty strings remain empty; missing values are omitted and wrong-type/null selections refuse.

registry.npmjs.org receives the chosen name/version or query and request metadata. One metadata request can read a whole packument; it never downloads packages or executes scripts. Results do not establish security, installability, ownership, completeness or freshness. Request headers are fixed (Accept: application/json, Accept-Encoding: identity, User-Agent: Plaice-npm-public-v1/1.0), with no body, omitted credentials, no-store cache and redirect refusal. Actual Response status 200, a nonempty exact final URL and no redirected response are required; returned URLs are never followed.

The exposed decoded response is capped at 1 MiB and 8,192 read iterations, with bounded UTF-8, original JSON duplicate-key validation, depth 32, nodes 32,768 and selected output 65,536 bytes. The shared policy accepts at most five known content codings; identity stands alone. For absent/identity coding, canonical Content-Length must match exposed bytes; compressed wire length is not compared to decoded bytes and no second decompression occurs. Numeric tokens are converted with Number only for finite classification, then the original opaque lexeme is retained; selected search totals require an original canonical unsigned decimal safe integer before a separate projection conversion. Fractions, exponent forms and negative zero are refused, rather than rounded into authority.

One attempt applies even when block retries are set to three. Provider and gateway each use 10-second timers after their respective preparation boundaries; there is no preparation, whole-invocation or cleanup deadline. The final trusted check follows both dispatch callbacks and the prepared-account hook, revalidating the exact public operation, current actor/workspace/live state and run authority. Persistence reads fail closed on errors or missing required records, without the legacy memory fallback. Observations are not atomic with an external send. Errors show only npm registry read unavailable. for provider failures. Matching-revision Review, human confirmation, actual organization/Prisma persistence, native Fetch/DNS/redirect/decompression behavior and live provider compatibility remain unverified by the local synthetic tests.

Local memory executions are saved only when they finish. A missing saved memory run cannot establish noncancellation: dispatch instead preserves the runtime-generated run identity captured before callbacks, current local scope/live checks and AbortSignal. A matching saved memory record, when present, must also remain in scope and not canceled. Persistence mode requires its authoritative run record and never substitutes this memory limitation.

Zulip and Google Books selected metadata

Save an existing Zulip Basic credential with exactly apiOrigin, email and apiKey, then select that exact zulipBasicV1Api account on the block. The origin is the exact public root HTTPS origin, with no port or path; the email is a personal identifier used with the API key. There is no discovery or connection probe. Google Books uses its separate googleBooksApiKey account with only apiKey; the key identifies the application for public reads and uses project quota. Neither profile imports historical credentials, OAuth or a service-account key. Save makes zero provider requests and leaves identity, privileges, expiry, quota and access unverified; Test is unavailable. Original malformed/duplicate-key Advanced JSON is refused before Format JSON or Use fields can silently collapse it, while generic nonempty persistence may still store material refused at execution.

Both profiles locally accept 1–2048 visible ASCII key characters in received JSON up to 32768 JavaScript units or 65536 UTF-8 bytes; these are local support limits, not provider credential-format guarantees. Keep keys and authenticated request URLs out of workflow input and output. Enable the workspace live control for Zulip or Google Books; when no explicit setting exists, WORKFLOW_LIVE_ZULIP_BASIC_V1 and WORKFLOW_LIVE_GOOGLE_BOOKS_API_KEY provide the respective defaults. Saving an account does not confirm execution or grant rights.

Every disclosure switch starts unchecked and requires an actual boolean true, followed by independent execution confirmation. Search terms can disclose interests or personal identifiers. A saved setting or an automatically approving callback is not a person's consent. Select existing accounts explicitly; no account or identity lookup is added. Current account, actor, live setting and run authority must remain valid through final dispatch. These observations do not create an atomic lock across an external send.

Zulip IDs are exact positive decimal strings, locally at most 32 digits; original numeric response identities are projected losslessly, including above JavaScript's safe integer range. This is not a full vendor int64/range guarantee. Optional selected text can retain explicit null/empty locally, actual flags preserve false, and a related creator ID can retain original zero as text. No messages, subscriptions, list/history, automatic pages or writes are provided.

Google Books volume IDs are opaque 1–128-character strings, not ISBN lookup inputs. Search preserves its exact full-text query and operators, locally 1–512 JavaScript units or 2048 UTF-8 bytes without controls or unpaired surrogates. Supply explicit integer startIndex 0–1000000 and maxResults 1–20; there is no automatic next page or all-results claim. Optional filter is ebooks/free-ebooks/full/paid-ebooks/partial, printType all/books/magazines, orderBy newest/relevance. Totals are original canonical nonnegative integer JSON tokens up to 32 digits, returned as text. A total is not a stable snapshot or continuation guarantee. Page count accepts an original canonical integer 0–10000000 before numeric conversion; rounding-prone fractions, exponents and negative zero are refused. Missing, null, zero, empty authors and author order/duplicates remain distinct only where supported by this local projection.

The Google key is carried internally in the fixed query, while Zulip uses Basic Authorization. Requests and exact Response URLs remain internal. No returned URL is followed or returned. Books descriptions contain HTML and are omitted, along with preview/download links, thumbnails, sale/access/user information and content. A free/full/partial filter does not establish rights, geographic access or availability. No library, OAuth, refresh, service-account import, full text, preview, download or write capability is added.

Each operation makes one attempt even when block retries are higher, with no redirects, arbitrary headers/hosts, fallback requests or automatic paging. The provider's 10-second and gateway's 11-second deadlines begin after confirmation and credential preparation; preparation and blocked callbacks have no whole-invocation deadline guarantee. Decoded bodies are bounded to 1 MiB with fatal UTF-8 decoding, original duplicate-aware JSON, finite parser/read/node/depth/string/number/work/projection budgets, and supported content codings without wire-length equality. Known credential fragments are checked in actual keys/string values, including omitted body fields and new public metadata, without fabricating array-index or zero/false strings. This is not arbitrary unknown-secret or PII detection.

Local synthetic checks do not verify live vendor compatibility, account permissions, quota/billing/geography, actual DNS/redirect/decompression behavior, organization/Prisma persistence, human consent or a matching-revision Review walkthrough. Review those limits before relying on an execution. Saving or viewing metadata does not certify freshness, completeness, ownership or operational success.

Google BigQuery and Business Profile manual metadata reads

Save the exact native manual profile and select that saved account on its matching block. BigQuery uses googleBigQueryManualBearerApi with only accessToken. Business Profile uses googleBusinessProfileManualBearerApi with exactly accessToken, accountName and locationName. Enter the existing bearer token without a Bearer prefix or surrounding whitespace. Business destinations must be exact accounts/... and locations/... strings: each suffix starts with an ASCII letter or digit, followed by at most 255 ASCII letters, digits, underscores or hyphens. Leading zeros remain strings. These destinations do not prove ownership, grants or an account/location association.

Save performs local validation only, with zero provider requests and an unverified result. Test, OAuth acquisition, refresh, service-account import and automatic conversion of historical credentials are unavailable. Project approval, Business Profile eligibility, IAM, scopes, token expiry, API access, quotas and charges remain unverified. Saving grants no execution authority; all four operations require independent execution confirmation, with no self-approval or acknowledgment input.

BigQuery's bounded native ID profile accepts 1–128 ASCII letters, digits, underscores or hyphens for projectId; 1–1024 ASCII letters, digits or underscores for datasetId; and 1–1024 ASCII letters, digits, underscores, hyphens or spaces for tableId. Table spaces and leading zeros are preserved. Project/dataset controls use ordinary whitespace normalization; supply exact IDs without surrounding whitespace. Whole expressions must resolve to strings for IDs and actual numbers for page sizes; numeric strings are refused. Neither profile accepts extra input keys. locations.get cannot override the saved account or location.

Continuation inputs are optional nonempty printable-ASCII strings of at most 2048 characters, preserving whitespace. Use a returned token only in a later explicit invocation for the same selected account and page size, and the same BigQuery project where applicable. Token history and applicability are not established locally. Omission, null, empty strings, false and zero remain distinct where the projection supports them; an empty or null returned continuation is not a valid continuation input. BigQuery preserves omitted dataset collections and unreachable locations; neither proves project completeness. Business Profile returns an empty locations array when that collection is omitted, always keeps accountComplete:false, and derives hasMore from a nonempty returned token. Its fixed read mask is name,title,storeCode,languageCode,labels,openInfo; other returned fields are omitted.

Use the corresponding workspace live control. An explicit workspace boolean overrides the environment. Without that boolean, only the exact environment string false disables WORKFLOW_LIVE_GOOGLE_BIGQUERY_MANUAL_BEARER or WORKFLOW_LIVE_GOOGLE_BUSINESS_PROFILE_MANUAL_BEARER; an absent flag enables under the shared policy. Account, actor, run, confirmation and live checks still apply at dispatch.

Local response limits include a 1 MiB exposed decoded body, 8192 reader results including the terminal result, JSON depth 32, 32768 JSON values and a 256 KiB serialized selected result. These are local parser/projection limits, not raw-wire or decompression guarantees. Oversized, malformed or secret-colliding selected data refuses the whole result rather than returning a truncated success. BigQuery schema projection permits at most 256 field entries across nesting, with a nesting-depth check of 8; selected field text is at most 2048 UTF-16 units. Its quoted counters/timestamps permit up to 128 decimal digits. Business Profile permits at most 100 labels of 2048 UTF-16 units each, selected text up to 2048 units, and openInfo.status up to 256 units with optional nullable boolean canReopen.

Each operation uses one request attempt to its fixed HTTPS origin, even with block retries enabled, with no redirects, returned-URL following or automatic continuation. The provider's 10-second execution budget and gateway's 11-second budget start after preparation; they do not bound credential or confirmation preparation. Inputs and selected results may enter ordinary retained workflow configuration, run history and downstream steps. BigQuery does not execute SQL, jobs, table-data reads, uploads, exports, writes or billing-project overrides. Business Profile does not expose reviews, posts, writes, arbitrary masks or discovery. These four operations extend the preceding 141-action group to 145; native service coverage is partial, with zero n8n import aliases and no action/field parity. Local simulated coverage does not establish real Google access, eligibility, billing, response compatibility, human confirmation or matching-revision Review behavior.

Google Chat, Storage and Natural Language manual tokens

Save an existing token with only accessToken using the exact native profile: googleChatManualBearerApi, googleCloudStorageManualBearerApi or googleNaturalLanguageManualBearerApi. Select that saved account on the matching block and enable its workspace live control. Save is local and unverified, with zero provider requests. Test, token acquisition, refresh and service-account import are unavailable. Saving does not establish identity, permissions, membership, expiry, quota, billing or execution authority. Existing OAuth credentials are not converted.

Google Chat space metadata

spaces.get requires an exact resourceName including spaces/; preserve opaque case and leading zeros. spaces.listPage requires numeric pageSize 1–100 and optionally the exact pageToken from a preceding page with the same size. Continuation whitespace is preserved. Both require independent execution confirmation, without an acknowledgment input.

Results select space name and supported optional display name, type, threading state and flags. Get checks the returned identity. A list returns one page with pageScope: single-page, accountComplete: false, hasMore and optional continuation. It does not establish complete account or organization coverage. User versus app identity, scopes, membership and Chat configuration remain unverified. Messages, membership changes, administrative access, search, uploads and webhooks are unsupported.

Google Cloud Storage metadata

buckets.getMetadata requires bucket. objects.listPage requires bucket and numeric maxResults 1–100, with optional exact prefix and pageToken. The supported bucket subset uses 3–63 lowercase ASCII characters, excludes dotted and reserved names, and accepts neither gs:// nor URLs. Prefixes preserve Unicode and empty text within 1024 UTF-8 bytes; malformed Unicode and line breaks are refused. Both operations require independent confirmation, without an acknowledgment input.

Only selected bucket/object metadata is returned, using a fixed noAcl projection. Object content, ACLs, custom metadata and returned links are excluded; no download follows. Each item must match the requested bucket and prefix. Quoted generation/metageneration remain decimal strings within signed 64-bit support bounds; size remains a decimal string within unsigned 64-bit bounds. No numeric rounding occurs. A page declares bucketComplete: false and pageScope: single-page. Continue explicitly with compatible prior settings; token history is not verified. Metadata does not prove content integrity, freshness, permissions or completeness. Writes, transfers, signed URLs and requester-pays billing-project selection are unsupported.

Google Cloud Natural Language sentiment

analyzeSentiment requires exact plain text and separate initially unchecked acknowledgeTextDisclosure and acknowledgePossibleBilling choices, followed by independent execution confirmation. Text is bounded to 5000 Unicode code points, 10000 UTF-16 units and 32768 serialized request bytes. One fixed POST sends document.type: PLAIN_TEXT and encodingType: UTF16. Markup and entities remain literal; URLs and Cloud Storage documents are not fetched.

Output selects documentSentiment with optional score/magnitude and optional language. Metrics remain original JSON numeric-token strings, including exponent spelling and negative zero. Local bounds are score −1 to 1 and magnitude 0 to 10000000; missing metrics stay absent. The analysis does not establish health, identity, correctness or safety. Google may have processed and billed a failed request; a local timeout or cancellation does not prove otherwise. The selected-output operation excludes entities and sentence results. HTML mode, source URIs and explicit language/model selection are unsupported. Current pricing is unverified.

All five operations use fixed HTTPS origins, one attempt even when block retries are enabled, no redirect following and no automatic pages. Provider 10-second and gateway 11-second budgets begin after preparation; they do not bound credential or confirmation preparation. Inputs, text, metadata and results may enter ordinary retained workflow configuration, history and downstream steps. These operations extend the preceding 136-action group to 141. These five original selected-output profiles have no automatic n8n import aliases or full-node parity; the separate full-response compatibility profiles below have narrower explicit import support. Local app/control tests use simulated provider responses and confirmation; real Google permissions, compatibility, billing, human consent and matching-revision Review remain unverified.

Google Translate and Contacts manual tokens

Save an existing account containing only accessToken: select googleTranslateBearerV2Api for Google Translate · Basic v2 plain text, or googleContactsManualBearerApi for Google Contacts · selected reads. Select that exact saved account on its block. Save is local and unverified with zero provider requests; Test, token acquisition, refresh and service-account import are unavailable. Historical OAuth credentials are not converted. Rights, scopes, expiration and API enablement remain unverified.

Enable the corresponding workspace live control. When no explicit workspace setting exists, WORKFLOW_LIVE_GOOGLE_TRANSLATE_BEARER_V2 and WORKFLOW_LIVE_GOOGLE_CONTACTS_MANUAL_BEARER supply the defaults. Saving an account grants no execution authority. Current account, actor, run and live settings must remain valid through dispatch; this is not an atomic lock across an external request.

Translate sends one scalar JSON request to the fixed Google Basic v2 endpoint with format:text. Whitespace, markup, entities and emoji remain literal text; the adapter neither decodes entities nor renders HTML. Local limits are 5000 Unicode code points and 32768 serialized request bytes. Language syntax is checked locally; Google language support is unverified. The two disclosure and possible-billing switches start unchecked and do not replace independent execution confirmation. Google may process or charge for an attempted request even when its final result is unavailable. Inspect an uncertain translation before deliberately repeating it.

Contacts requires independent confirmation, with no additional acknowledgment input. Keep resource names as opaque strings, including case and leading zeros; people/me and suffix-only IDs are refused. Choose names, emailAddresses, phoneNumbers, organizations or birthdays; wildcards are unavailable. Each selected field permits at most 20 returned entries. Only supported selected subfields are returned; metadata, etags, totals and photo URLs are omitted. Missing values, null, false, zero, order and duplicates remain distinct where supported by the projection. Get checks the returned identity. One page never establishes complete account coverage. Use a returned token only in a later explicit invocation with the same mask and page size; the adapter does not bind earlier request context or validate token history. No automatic paging, search, warmup, sync, photo fetches or writes are provided.

Both profiles make one attempt, with fixed HTTPS origins, no redirect following and no automatic retry. The 10-second provider and 11-second gateway budgets begin after host preparation. Decoded responses are capped at 1 MiB and selected output at 256 KiB. These are local support limits, not provider compatibility or quota guarantees. Source text, contact fields and results can enter ordinary workflow configuration, retained run history and downstream steps. These three operations extend the preceding 133-action group to 136; ordinary and Tool coverage remains partial. These three selected-output operations have no automatic n8n import aliases or full-node parity; the additive full-response language profiles below have explicit limited import support.

Local tests use synthetic provider, confirmation and UI responses. Live Google token validity, JSON acceptance, selected fields, rights, billing and quotas remain untested. Authenticated account Save, human confirmation, organization/Prisma persistence and the changed Help steps still require a matching-revision Review walkthrough.

Microsoft OneDrive and SharePoint selected metadata reads

Choose the matching manual account in Integrations and select that exact saved account on the workflow block. microsoftOneDriveManualBearerApi contains exactly accessToken and driveId; microsoftSharePointManualBearerApi contains exactly accessToken and siteId. Supply an already-acquired Microsoft Graph bearer token without a Bearer prefix. The selected drive or site stays in the credential profile, outside operation input. Save is local and unverified with zero provider requests; Test, OAuth acquisition, refresh, tenant-consent setup and historical credential conversion are unavailable. Saving does not establish token audience, permissions, expiry, tenant consent, account eligibility, drive/site ownership or execution authority.

OneDrive drive and item IDs use the native bounded profile: 1–256 ASCII letters, digits, underscores, hyphens or exclamation marks. Preserve case and leading zeros. The item root alias, resource paths and URLs are refused. SharePoint site ID uses the supported global .sharepoint.com hostname followed by two hyphenated hexadecimal GUIDs, comma-delimited (hostname,GUID,GUID); preserve case and zeros. This is a supported native subset, not Microsoft's complete ID grammar or a sovereign-cloud promise. No per-call drive/site override, discovery or extra input keys are accepted.

Whole expressions must resolve to strings for IDs and actual numbers for page sizes; numeric strings are refused. All four reads use fixed Microsoft Graph v1.0 routes at https://graph.microsoft.com. OneDrive selects only id,name,size,createdDateTime,lastModifiedDateTime. Its size retains the original nonnegative Int64 JSON integer as decimal text, including "0", without converting it through an imprecise JavaScript number. Omitted fields stay omitted and explicit null stays null. SharePoint site selection is id,name,displayName,description,createdDateTime,lastModifiedDateTime,isPersonalSite; list selection is id,name,displayName,description,createdDateTime,lastModifiedDateTime,list, whose projected facet contains only hidden, contentTypesEnabled and template. Supported booleans retain false and null; empty strings, omission, provider order and duplicate list entries remain distinct. These projections do not prove Microsoft will emit every admitted nullable shape.

Both collections require a real value array in the received response; missing or null collections are refused rather than fabricated as empty success. Each page read exposes only its first explicit page. A bounded nonempty @odata.nextLink sets the respective hasNextPage or hasNext boolean, but its URL is discarded, never returned and never followed. There is no continuation input. A false boolean does not establish account, drive, site or list completeness; SharePoint system lists are omitted by default. OneDrive always reports accountComplete:false on children pages. Do not use these operations for an export or complete inventory.

Review independent execution confirmation for every operation; no self-approval or acknowledgment field is added. Use the matching workspace live control. An explicit workspace boolean overrides the environment; without an override, only exact false disables WORKFLOW_LIVE_MICROSOFT_ONEDRIVE_MANUAL_BEARER or WORKFLOW_LIVE_MICROSOFT_SHAREPOINT_MANUAL_BEARER, and an absent flag enables under the shared policy. Current account, actor, run and confirmation checks still apply at dispatch. Inputs and selected results may enter ordinary retained workflow configuration, run history and downstream steps.

Each operation makes one request attempt with no redirect, retry, returned-link following or automatic page. The finite carrier limits the exposed decoded body to 1 MiB and selected serialized output to 256 KiB, with existing JSON/read budgets. These are local decoded-body/parser limits, not raw-wire or decompression guarantees. The provider's 10-second budget and gateway's 11-second budget begin after preparation; they do not bound credential or confirmation preparation. Malformed, oversized or secret-colliding responses refuse the whole result.

No content, download, file/item content, writes, sharing, search, trigger, automatic paging, token lifecycle or n8n import alias is included. Ordinary and Tool coverage is partial service coverage without action or field parity. The shipped SharePoint node uses SharePoint REST, while these native reads use Microsoft Graph; site.get is a new native operation, not a shipped SharePoint site action. Static shipped-source evidence and local synthetic tests do not verify human approval, Review behavior, real token scopes/audience, permissions, tenant consent, drive/site access or the exact encoded query and provider response. The four operations extend the preceding 145-action group to 149.

Google Slides selected metadata reads

Choose Google Slides · selected metadata reads (google-slides-metadata) for this separate manual-token profile. The existing Google Slides OAuth/service-account integration and its eight actions stay separate. Save googleSlidesManualBearerApi with only accessToken, then select that exact saved account on the block. Supply an already-acquired bearer token without a Bearer prefix. Save is local and unverified with zero provider requests; Test, token acquisition, refresh and service-account signing are unavailable. Saving establishes neither account access, scopes, expiry nor execution authority. Google quota applies.

Select a nonempty unique array from the operation's choices. Whole expressions must produce actual string arrays, not JSON text. IDs use the native subset of 1–256 ASCII letters, digits, underscores or hyphens; preserve case and leading zeros. URLs are refused. These are local support limits, not Google's complete ID grammar. Returned identity must match the exact requested ID.

The response includes the identity and only requested metadata. slides permits at most 100 summaries; a larger response is refused rather than truncated. Each summary contains objectId and supported pageType, revisionId and slideProperties; the latter projects only layoutObjectId, masterObjectId and isSkipped. Missing fields, null and false remain distinct where supported. No full slide content, notes, comments, thumbnails, media downloads, writes or automatic follow-up is included.

Independent runtime confirmation remains required, without an extra acknowledgment input. WORKFLOW_LIVE_GOOGLE_SLIDES_METADATA_MANUAL_BEARER controls this provider: an explicit workspace boolean overrides the environment; otherwise exact false disables and an absent flag enables. IDs go to Google, and inputs, selected results and history may persist normally and reach downstream steps. The two reads provide partial service coverage, without n8n field/item/error parity, import aliases or credential conversion. Local synthetic checks do not verify real Google access, quota, human consent or Review behavior.

Microsoft Entra ID selected metadata reads

Choose Microsoft Entra ID · selected metadata reads (microsoft-entra). Save microsoftEntraManualBearerApi with only an existing accessToken, without a Bearer prefix, then select that exact saved account on the matching block. Save is local and unverified with zero provider requests; Test is unavailable. Saving does not establish identity, permissions, audience, expiry, tenant consent, privacy consent or execution authority. Token acquisition, refresh, OAuth consent and n8n credential conversion are unavailable.

Get accepts only a GUID-shaped existing ID in this native subset: hexadecimal 8-4-4-4-12 groups separated by hyphens. Preserve exact case and leading zeros. Microsoft Graph IDs are opaque, not universally GUIDs; UPNs, email addresses, URLs and other ID shapes are outside this input profile. A differently serialized response ID is refused, without normalization. Numeric strings are not page sizes. Extra input keys and arbitrary metadata fields are refused.

The page is always an incomplete inventory. Continuation is discarded, never returned or followed, and no later-page input is supported. hasNext:false does not prove complete directory coverage. Selected missing fields, null, false, empty strings, order and duplicates remain distinct where supported. There is no discovery, search, write, membership change or automatic paging.

Independent runtime confirmation remains required, without an extra acknowledgment input. WORKFLOW_LIVE_MICROSOFT_ENTRA_MANUAL_BEARER controls this provider: an explicit workspace boolean overrides the environment; otherwise exact false disables and an absent flag enables. Personal and directory metadata may persist in ordinary workflow inputs, results, history and downstream steps. Permissions, account eligibility and provider charges remain unverified. These four reads provide partial ordinary/Tool service coverage, without n8n locator/field/item/error parity or import aliases. Static shipped-source evidence and local synthetic checks do not prove live Graph access, human consent or Review behavior.

Microsoft Graph Security selected profile reads

Choose Microsoft Graph Security · selected profile reads (microsoft-graph-security). Save microsoftGraphSecurityManualBearerApi with exactly accessToken, an already-acquired token without a Bearer prefix. The form labels it Existing unverified Microsoft Graph Security access token.

Save the exact matching manual account locally and unverified, then select that saved account on the matching block. Save account sends zero provider HTTP requests; Test connection is unavailable. Existing credential disclosure choices remain separate from independent per-run execution confirmation. Saving grants no execution authority and adds no operation acknowledgment field. Token identity, audience, expiry, permissions, customer eligibility, quota, licensing and any charges remain unverified. Inputs, projected outputs, downstream data and run history may persist in ordinary workflow storage. No OAuth acquisition, refresh, signing, discovery, search, writes, membership changes or n8n import aliases are provided.

The operation labels are Get selected security control-profile metadata and Read one security control-profile first page. The native ID subset is [A-Za-z0-9_-]{1,256}: preserve case and leading zeros, refuse whitespace rather than trimming, and supply neither URLs nor aliases. This is not the full opaque provider ID grammar. Graph applies $top per security provider; a response exceeding the requested cap is refused rather than truncated.

Requests use the fixed Microsoft Graph v1.0 /security/secureScoreControlProfiles endpoint. There is no $select query: the full upstream profile arrives at the bounded transport before local projection. The 1 MiB limit covers exposed decoded response bytes, not raw-wire size or decompression. Selected profile fields are id, optional bounded text actionType, controlCategory, implementationCost, lastModifiedDateTime, service, tier, userImpact, and optional nullable boolean deprecated. Output discards content, comments, tenant IDs, URLs, vendor information and nested payloads; those discarded fields still reach the transport. Selected absent, null, empty text and false values remain distinct where supported. Continuation is discarded, never returned or followed; hasNext:false never proves a complete inventory.

The live control is WORKFLOW_LIVE_MICROSOFT_GRAPH_SECURITY_MANUAL_BEARER. Explicit workspace booleans override the environment; otherwise only exact false disables and an absent flag enables. This two-read manual profile is narrower than the original n8n node's secure-score reads and control-profile get/getAll/update actions. It adds no security scores, alerts, profile updates, comments, content downloads or full n8n action/field parity.

Google Workspace Directory metadata reads

Choose Google Workspace Directory · metadata reads (google-workspace-directory). Save googleWorkspaceDirectoryManualBearerApi with exactly accessToken and customerId. The form labels are Existing Google Workspace Directory access token and Selected Google Workspace customer ID. The customer destination belongs to the account, outside operation input.

Save the exact matching manual account locally and unverified, then select that saved account on the matching block. Save account sends zero provider HTTP requests; Test connection is unavailable. Existing credential disclosure choices remain separate from independent per-run execution confirmation. Saving grants no execution authority and adds no operation acknowledgment field. Token identity, audience, expiry, permissions, customer eligibility, quota, licensing and any charges remain unverified. Inputs, projected outputs, downstream data and run history may persist in ordinary workflow storage. No OAuth acquisition, refresh, signing, discovery, search, writes, membership changes or n8n import aliases are provided.

The native subset for resource and saved customer IDs is [A-Za-z0-9_-]{1,128}. Preserve exact case and leading zeros; raw whitespace is preserved and refused rather than silently trimmed. Email, domain, URL and my_customer customer aliases are unavailable. This bounded subset is not Google's complete opaque ID grammar. The read labels are Get user metadata, Read first user metadata page, Get group metadata and Read first group metadata page.

Requests use fixed https://admin.googleapis.com/admin/directory/v1 user/group paths with selected fields; lists send the exact saved customer and users use projection:basic. Returned users must echo the selected customer exactly. Group responses expose no customer field: group-customer ownership cannot be independently verified from these responses, including a successful get or first page. Selected bearer/account/query binding is not proof of group ownership.

directMembersCount is a quoted 1–19-digit decimal string at most 9223372036854775807; zero and leading zeros are preserved. Numeric JSON counters are refused. Supported missing, null and false remain distinct. An omitted collection becomes empty; explicit null or an over-cap collection refuses the response. Lists preserve order and duplicates. Both first-page operations always report inventoryComplete:false, including empty pages and hasNext:false. Only hasNext survives; no continuation input is accepted and no token is returned or followed. Names, emails, password payloads and unselected fields are discarded after full-body secret scanning; ordinary retained metadata can still be sensitive.

The live control is WORKFLOW_LIVE_GOOGLE_WORKSPACE_DIRECTORY_MANUAL_BEARER, with the same explicit workspace override and exact-false/absent-enabled behavior. These four reads are a limited manual profile alongside the broader original Google Workspace Admin node, whose device/user/group management, writes, membership operations and OAuth lifecycle are not adopted.

Webex selected meeting metadata

Choose Webex · selected meeting metadata (webex-meeting-metadata). Save webexMeetingManualBearerApi with exactly accessToken, an existing unexpired user-level bearer token without a Bearer prefix. An eligible existing Webex Meetings/CI account and meeting:schedules_read rights are prerequisites; saving does not verify them.

Select the exact saved account on its matching block. Save account is local and unverified, with zero provider HTTP; Test connection is unavailable. Existing credential disclosures do not replace independent per-run confirmation, and no operation acknowledgment input is added. Saving grants no execution authority. Inputs, selected results, history and downstream data use ordinary workflow retention. Full bounded upstream responses can contain private fields in adapter memory before projection. Current token rights, expiry, quota, provider eligibility and live compatibility remain unknown.

Meeting IDs use the restricted ASCII subset [A-Za-z0-9_-]{1,512}, preserving case and leading zeros. This is narrower than Webex's opaque-string grammar; whitespace, URLs, slashes, percent escapes and other ID forms are refused. Get requires an exact response-ID match without normalization.

Both list dates must be calendar-valid UTC seconds in YYYY-MM-DDTHH:mm:ssZ form, with inclusive from, exclusive to and a positive range of at most 31 days. List sends explicit max, from and to only. Get leaves current at its false default; List leaves current at true and meetingType at meetingSeries. Series overlap the range; returned occurrence start/end timestamps can fall outside it. No current or meeting-type query override is offered.

Missing optional metadata is omitted. Present null, false, empty or invalid selected values are refused rather than defaulted; timestamps support valid UTC seconds with optional 1–3 fractional digits. continuationPresent means only that a nonempty Link header exists, including a link with another relation: it is not parsed next-page proof. No link is returned or followed, and inventory is always incomplete, including empty pages and a false flag. Names, agenda, passwords, join links, email, participants, content, transcripts and recordings are not returned; no meeting/message writes, impersonation, OAuth acquisition or refresh are provided.

The live control is WORKFLOW_LIVE_WEBEX_MEETING_MANUAL_BEARER. An explicit workspace boolean overrides the environment; otherwise only exact false disables and unset enables. Selected-account authority and live permission are rechecked at later dispatch boundaries. One GET attempt can consume provider quota, with a 10-second provider deadline and a 262,144-byte decoded-body cap, not a raw-wire cap. No paging or retries occur. Ordinary/Tool service coverage is partial, without n8n import aliases, credential conversion or full action/field/OAuth parity.

SecurityScorecard selected company reads

Choose SecurityScorecard · selected company reads (securityscorecard-read). Save securityScorecardReadApi with exactly apiKey, an existing API key. The selected key must already have access to the existing scored company or portfolio. A terminal 403/404 never creates or enrolls a company, adds it to a portfolio, generates a scorecard or upgrades a plan.

Select the exact saved account on its matching block. Save account is local and unverified, with zero provider HTTP; Test connection is unavailable. Existing credential disclosures do not replace independent per-run confirmation, and no operation acknowledgment input is added. Saving grants no execution authority. Inputs, selected results, history and downstream data use ordinary workflow retention. Full bounded upstream responses can contain private fields in adapter memory before projection. Current token rights, expiry, quota, provider eligibility and live compatibility remain unknown.

The domain is at most 253 characters, with labels at most 63 characters and an alphabetic start to the final label. Uppercase, whitespace, URL/path/port, wildcard, underscore and custom scorecard identifiers are refused without normalization. Summary grade is one of A, B, C, D, F or ?; absent score stays absent while calculating, and present null is refused. Company name is nonempty bounded text at most 512 characters.

At most 100 unique factors are accepted. Each has required ASCII letters/digits/underscore/hyphen name of 1–256 characters, optional integer score 0–100 and optional supported grade. issueCount is derived by summing the counts of at most 100 issue_summary entries per factor, bounded to 10,000,000. Absent issue_summary omits issueCount; an empty array yields zero; null or invalid selected fields refuse the response. Required total is an integer 0–10,000,000 no smaller than the returned factor count. No issue evidence, IPs, reports, downloads, returned links or arbitrary provider fields are exposed; completeness is unverified.

The live control is WORKFLOW_LIVE_SECURITYSCORECARD_READ, with explicit workspace boolean overrides, exact-false disabling and unset enabled. Selected-account authority and live permission are rechecked at later dispatch boundaries. One GET attempt may consume quota, under a 10-second provider deadline and 262,144 decoded response bytes. No retries, paging, writes, OAuth lifecycle or full n8n action/field/import parity is included. Ordinary/Tool inventory coverage means only these two company reads.

Wise profile metadata reads

Choose Wise · profile metadata reads (wise-profile-read). Save wiseProfileReadApi with exactly apiToken, an existing eligible Wise Business personal API token without a Bearer prefix. Eligibility, token permissions and compatibility are unknown. This fixed profile uses https://api.wise.com/2026Q4, deliberately different from shipped legacy TransferWise v1, with no fallback or configurable origin.

Select the exact saved account on its matching block. Save account is local and unverified, with zero provider HTTP; Test connection is unavailable. Existing credential disclosures do not replace independent per-run confirmation, and no operation acknowledgment input is added. Saving grants no execution authority. Inputs, selected results, history and downstream data use ordinary workflow retention. Full bounded upstream responses can contain private fields in adapter memory before projection. Current token rights, expiry, quota, provider eligibility and live compatibility remain unknown.

profileId is positive canonical int64 decimal text up to 9223372036854775807: no zero, leading zeros, signs, whitespace, fractions, exponents or input numbers. Preserve exact IDs above 2^53, for example "9007199254740993"; never convert them to a JavaScript number. Original numeric response tokens become exact decimal strings and Get requires exact identity. This is a deliberate subset of Wise int64 IDs.

type is PERSONAL or BUSINESS. currentState, when present, is HIDDEN, VISIBLE or DEACTIVATED; absence remains omitted and present null or another value is refused. The upstream list must be a bare array with at most 100 entries, preserving order and duplicates. An empty list is valid; larger lists are refused rather than truncated. inventoryComplete:false never certifies a complete inventory. No paging input, continuation or automatic follow-up is supported.

Full bounded upstream profiles can include names, email, addresses, birth dates and contact details before local projection. Only id, type and optional currentState survive. No balances, statements, transfers, quotes, recipients, rates, money movement, writes, partner onboarding, SCA, private keys, impersonation, OAuth acquisition or refresh are available.

The live control is WORKFLOW_LIVE_WISE_PROFILE_READ, with explicit workspace boolean overrides, exact-false disabling and unset enabled. Selected-account authority and live permission are rechecked at later dispatch boundaries. One GET attempt may consume quota, with a 10-second provider deadline and 262,144 decoded response bytes. No retries or legacy-version fallback occur. The ordinary Wise inventory row records only partial service coverage; no Tool row is materialized, n8n import alias added or full original-node/action/field/OAuth parity claimed.

Travis CI selected hosted build metadata

Choose Travis CI · selected hosted build metadata and save an existing hosted .com v3 manual token as a workspace credential of type travisCiManualApiTokenV3Api, with only the required password field apiToken. The value preserves exact visible ASCII characters 0x21–0x7e, 1–4096 characters, without trimming. Select that exact saved account in the block. This profile fixes the origin to https://api.travis-ci.com/v3, uses Authorization: token … and Travis-API-Version: 3, and does not support a custom host or legacy .org service.

Build IDs use canonical positive decimal text from 1 through 9007199254740991; spaces, leading zeros, exponents and numeric coercion are refused. This is a conservative local subset, not a vendor maximum. A selected item contains id as exact decimal text and optional number, state and private. The response must supply a numeric ID and @type:"build"; the list root must have @type:"builds". number is 1–32 decimal digits, state matches [a-z][a-z0-9_ -]{0,63}, and private is boolean. Absent or null optional fields are omitted; invalid nonnull types refuse the response. Lists preserve upstream order and duplicates, refuse responses exceeding the requested limit and never establish complete inventory.

The fixed requests are GET /build/{buildId} and GET /builds?limit={limit}&offset=0. Original n8n build get and getAll have only these narrow metadata analogues. Cancel, restart, trigger, include options, sorting, full outputs and paging are unsupported. There are no original action/import aliases or automatic Tool wrappers.

Tapfiliate program affiliate metadata

Choose Tapfiliate · program affiliate metadata and save an existing user API key as a workspace credential of type tapfiliateManualApiKeyV16Api, with only the required password field apiKey. This key acts within one advertiser account; it is not read-only or program-restricted. It is sent unchanged as X-Api-Key to fixed https://api.tapfiliate.com/1.6. The value must contain 1–4096 visible ASCII characters 0x21–0x7e, without trimming. A matching quote pair around the entire value is refused; internal quotes are allowed.

Both IDs preserve case and use the conservative local [A-Za-z0-9_-]{1,128} subset, without normalization. Other vendor-valid forms may be unsupported. Each selected item requires string id; optional approved is omitted when absent, preserves false, and refuses null or nonboolean values. A matching affiliate ID does not independently verify program ownership.

Get uses GET /programs/{programId}/affiliates/{affiliateId}/; list uses exactly GET /programs/{programId}/affiliates/?page=1. There is no page-size input. The documented service default is 25; the local response cap is 100 items. Order and duplicates are preserved, and inventoryComplete is always false. No automatic paging, polling or returned-link follow is provided.

The original 12-action matrix remains incomplete: affiliate create/delete/get/getAll; affiliateMetadata add/remove/update; programAffiliate add/approve/disapprove/get/getAll. Only two analogous program-affiliate metadata subsets are available. Full action outputs, filters, options, writes and all-pages behavior remain unsupported; no old action/import aliases or Tool wrappers are added.

Travis CI and Tapfiliate account and run rules

Save and reopening send zero provider HTTP requests; accounts remain unverified and Test is unavailable. Save does not acquire, renew, introspect, validate or enable credentials, or grant execution authority. Use workspace credentials for ordinary workflows. Organization-scoped credentials are unsupported. Even in an organization-visible workspace, execution requires direct workspace membership, organization membership and execute authority; organization visibility alone is insufficient.

Existing credential disclosures remain separate from required independent per-run confirmation. There is no extra acknowledgment or billing input. Current selected-account, run and live authority are checked again before dispatch. Each operation attempts one GET, including after 429 or a transport failure even when workflow maxTries is 3. No retries, writes, polling, discovery, followed URLs or continuation are provided.

Each operation has a total 10-second deadline including preparation and projection, a 262144-byte decoded exposed body cap, 8192 reader calls, JSON depth 16 and 10000 nodes. These local bounds do not establish raw-wire or decompression confinement. The full bounded private upstream body enters memory before projection. Inputs, selected results, history and downstream data use ordinary workflow retention. A GET can consume quota; free quota, key rights, eligibility, expiry, applicable charges and access are not guaranteed.

The controls are WORKFLOW_LIVE_TRAVISCI_BUILD_METADATA and WORKFLOW_LIVE_TAPFILIATE_PROGRAM_AFFILIATE_METADATA. An explicit workspace boolean overrides the environment; otherwise only exact false disables and unset enables. See account fields and block inputs.

Onfleet selected team metadata

Choose Onfleet · selected team metadata (onfleet-teams-metadata). Save onfleetTeamsManualApi with exactly apiKey, your existing unrestricted Onfleet organization API key. Task-only scoped keys are ineligible. The adapter uses the key as the Basic username with an empty password; this is not OAuth. The key may permit broader vendor access than these two reads, and saving does not certify its rights or eligibility.

teamId is exactly 24 ASCII characters from [A-Za-z0-9_*~-], preserving case. This conservative subset refuses whitespace, URLs and other ID forms. Get requires the exact same response ID. The snapshot refuses more than 100 teams and duplicate IDs; a successful or empty snapshot never certifies complete inventory.

Output contains required id, optional nullable name, optional nullable timeCreatedMs and timeLastModifiedMs. Absent fields remain omitted and null remains null. A non-null name has 1–256 JavaScript text units and must pass the conservative label checks; unsafe names refuse the response rather than being shortened. Non-null timestamps preserve original nonnegative integer millisecond tokens as decimal text through 9007199254740991, without rounding or unit conversion. These optional fields are local policy, not a promise that every vendor response supplies them.

Only one GET is sent to https://onfleet.com/api/v2/teams or its exact team-ID path. No tasks, drivers, routing, managers, locations, writes, full inventory or legacy aliases are available. Original n8n Onfleet actions remain partially covered; the trigger is unassessed. See the shared account and run rules.

Storyblok EU published story metadata

Choose Storyblok EU · published story metadata (storyblok-published-eu). Save storyblokPublishedEuCdaV2Api with exactly apiKey, an existing EU public Content Delivery v2 token for a space you are allowed to access. Token type, space, location and rights are not introspected. The destination is fixed to https://api.storyblok.com/v2/cdn/stories with version=published; draft/private content, management API credentials and other regions are unsupported.

storyId is positive canonical decimal text through 9007199254740991; no numbers, zero, leading zeros, signs, whitespace, slugs, UUIDs or URLs. Get requires the exact response ID. List reads only page 1, with numeric integer limit 1–25, default 25. Duplicate IDs or UUIDs and responses larger than the selected limit refuse the whole result. ID expressions must resolve to strings; enter limit as a number in the editor.

The selected PublishedStory fields are required id, lowercase uuid, nonempty name up to 256 text units, createdAt, required nullable publishedAt and boolean isStartpage; updatedAt is optional nullable. Missing differs from null. The local timestamp subset requires a valid Gregorian date in years 0001–9999, T, seconds, optional 1–9 fractional digits and Z or a numeric offset; leap seconds and -00:00 are refused. Strings are preserved without timezone conversion.

Full upstream content, assets, links and component fields enter bounded memory before projection; selected output excludes body content, slugs and URLs. No management writes, publishing/unpublishing, automatic pagination, relation resolution or paging aliases are implemented. Original n8n content actions remain partial and management actions remain unported. See the shared account and run rules.

PostHog Cloud project metadata

Choose PostHog Cloud · project metadata (posthog-projects-metadata). Save posthogCloudProjectsPersonalApi with exactly personalApiKey, region and organizationId. Use your own existing personal automation key without Bearer, explicit exact lowercase us or eu, and your known lowercase organization UUID. No field is trimmed, defaulted, discovered or inferred. The saved region selects only https://us.posthog.com or https://eu.posthog.com; region and organization stay outside operation inputs.

Existing project:read scope, actual project visibility and organization/key policy must already permit access; detail requires at least MEMBER membership. Project-restricted personal keys are conservatively unsupported for listing. Do not broaden a key or bypass policy. This personal automation profile does not establish eligibility for apps other people install or connect, for which PostHog directs developers to OAuth.

projectId is positive canonical decimal text through 9007199254740991, without numbers, zero, leading zeros, signs, whitespace, UUIDs, URLs or @current. Get requires an exact ID match. List accepts numeric integer limit 1–25, default 25, with offset fixed to 0; duplicate IDs and oversized lists are refused. Every result must match the saved organization. ID expressions must return strings; enter the list limit as a number.

Six fields are locally required and non-null: id as exact decimal text, organizationId, lowercase uuid, name (1–200 text units), boolean isDemo and timezone (1–64 text units). False is preserved. Text is not trimmed; timezone is not certified as an IANA identifier. These are conservative local response requirements; formal vendor requiredness is unverified. Lists check ID uniqueness, without promising UUID uniqueness.

One GET targets /api/organizations/{organizationId}/projects/{projectId}/ or /api/organizations/{organizationId}/projects/?limit={limit}&offset=0. Raw upstream tokens, secrets and settings can enter bounded memory before projection. No queries, analytics, recordings, event capture, persons, writes, token management, billing or followed URLs are provided. All five original n8n write actions—alias.create, event.create, identity.create, track.page and track.screen—remain entirely unported. These distinct metadata reads do not stand in for those actions. See the shared account and run rules.

Onfleet, Storyblok and PostHog account and run rules

In Settings → Integrations, select the exact provider and save its required fields. Save and reopening send zero provider HTTP, leave the account unverified, and do not approve execution; Test is unavailable. Use workspace credentials for ordinary workflows. Organization-scoped credentials are unsupported. Even in an organization-visible workspace, executing these reads requires direct workspace membership, organization membership and current execute authority; organization membership alone is insufficient.

Select that exact saved account in the block. Existing credential disclosures remain separate from required independent confirmation for every run. There is no extra operation acknowledgment or billing choice. Account changes, lost access, cancellation and disabled live execution are rechecked before dispatch. No OAuth acquisition, renewal, token introspection or certification of rights, expiry or eligibility is supplied.

Each read makes one GET attempt with no retries, automatic paging, polling, redirects or followed links. A request can consume provider quota; no charge, refund, rights or expiry guarantee is made. Full upstream private fields enter bounded process memory before selected output is produced. Inputs, selected results, history and downstream data follow ordinary workflow retention; projection does not guarantee temporary-only storage or secret-free process memory.

Each provider has its existing 10-second deadline, 262,144 decoded exposed response-byte cap and 8,192 read-iteration bound. These are local support limits, not raw-wire, decompression, chunk-allocation or native-process confinement promises. Unsupported response shapes refuse as a whole.

The controls are WORKFLOW_LIVE_ONFLEET_TEAMS_METADATA, WORKFLOW_LIVE_STORYBLOK_PUBLISHED_EU and WORKFLOW_LIVE_POSTHOG_PROJECTS_METADATA. An explicit workspace boolean overrides the environment; otherwise only exact false disables and unset enables. Ordinary and Tool inventory rows record partial service coverage only: no original action/import aliases, credential conversion, Tool wrappers or full field/item/error compatibility is added.

Paddle Billing and Plivo account and run rules

Paddle Billing uses the exact saved live or sandbox environment, fixed https://api.paddle.com or https://sandbox-api.paddle.com, Bearer authentication and Paddle-Version: 1. Only one selected product or one finite product page is read. Existing product.read rights are required and unverified. There is no credential acquisition, rotation, account change, automatic paging or Classic fallback.

Plivo uses fixed POST https://api.plivo.com/v1/Account/{authId}/Message/, Basic authentication and JSON {src:from,dst:to,text,type:"sms"}. This is an external write with possible charges. Review the exact sender, single recipient and text independently for each run. The local queued result is not delivery. Save stays unverified and does not authorize sending; Test is unsupported. There is no new billing setting or acknowledgment. Plivo logging defaults to true; account/sender/region eligibility, quota, rights, retention, recipient consent and delivery remain unverified. No automatic retry, readback, callback or fan-out is provided. After invocation, failure or cancellation can leave an accepted request; check Plivo separately before considering another send.

Use an existing workspace credential and select that exact account. Organization-scoped credentials are unsupported. Execution requires direct workspace membership, organization membership and execute authority; organization visibility alone is insufficient. Credential disclosure is separate from required independent confirmation for each run.

Use workspace credentials for ordinary workflows. Organization-scoped credentials are unsupported. The live controls are WORKFLOW_LIVE_PADDLE_BILLING_METADATA and WORKFLOW_LIVE_PLIVO. An explicit workspace boolean overrides the environment; otherwise only exact false disables and unset enables. Current execution authority and live permission must permit dispatch. Each operation still requires independent confirmation regardless of provider defaults.

The full bounded upstream body enters server memory before projection. Inputs, selected results, workflow configuration, history and downstream data may be retained. Rights, permissions, expiry, account eligibility, quota and charges remain unverified. One request can consume quota. Local field, output and response limits can reject otherwise valid vendor responses. See exact credential fields and block fields and limits. Original Paddle Classic actions, Plivo MMS/calls and import conversion remain outside this scope.

ProfitWell Metrics and Peekalink Preview account and run rules

Use an existing workspace credential and select that exact saved account. Organization-scoped credentials are unsupported. Execution requires direct workspace membership, organization membership and execute authority; organization visibility alone is insufficient. Credential disclosure remains separate from independent confirmation of every run. Saving and reopening send zero provider HTTP, leave the account unverified and grant no execution authority; Test is unsupported.

ProfitWell fixes https://api.profitwell.com and makes one GET for settings, daily metrics, monthly metrics or plans. Its existing private token is the raw Authorization value; current descriptive request examples disagree with the machine Bearer declaration. No fallback is attempted. Daily requests must use the current or previous UTC month, checked again before dispatch; UTC is local policy. Monthly history enters bounded memory in full before local newest-month selection; no server history limit, start/end range or paging is invented.

Peekalink makes one POST to https://api.peekalink.io/ with JSON {link:url} and current Bearer authentication. Use only a public HTTPS link without credentials or sensitive query/path data. Local validation rejects known secret parameters and selected credential encodings, but cannot discover every arbitrary query secret, encrypted value or nested encoding. It does not perform DNS validation or certify provider fetch/redirect safety. The provider may fetch, redirect, cache and store preview/media data and consume quota or bill the account. This integration fetches or renders no returned media. After invocation, timeout, cancellation, invalid response or any other failure leaves processing/billing uncertain; do not automatically repeat the request.

Each operation requires its own independent run confirmation. Live controls are WORKFLOW_LIVE_PROFITWELL_METRICS and WORKFLOW_LIVE_PEEKALINK_PREVIEW: explicit workspace booleans override the environment; otherwise exact false disables and unset enables. Current selected-account and execution authority must remain valid at dispatch. Target-bound personal use has not been tested.

Full bounded upstream bodies enter server memory before selected projection. Inputs, selected financial or preview results, workflow history and downstream data use ordinary retention. Rights, expiry, account eligibility, quota and charges remain unverified. These selected subsets do not provide full n8n action, field, item-loop, raw-output or error compatibility, import aliases or automatic Tool materialization.

Local limits can refuse otherwise valid vendor data. Neither key acceptance nor provider permissions, financial completeness, URL processing, cost or human confirmation has been certified against a real provider. See credential fields and block fields.

Phantombuster account, authority and execution

Choose Phantombuster (phantombuster-agent-metadata) and an existing workspace account of type phantombusterMetadataApi. Its required apiKey is a password control with exact visible ASCII text, 1–4096 characters; do not trim it. Optional vendorOrganizationId accepts 1–128 ASCII letters, digits, underscores or hyphens. This selects the vendor organization using X-Phantombuster-Org; it grants no Plaice organization or workspace authority. Requests use endpoint-specific X-Phantombuster-Key, without old Key-1/v1 fallback.

Saving and reopening are local, send zero provider HTTP and leave the key unverified. Test is unsupported. Select that exact saved workspace account for a run; organization-stored credentials are unsupported. Execution requires direct workspace membership, organization membership and execute authority. Current-account revision and exact prepared-request binding are checked again before dispatch. Target-bound personal execution has not been tested. Credential disclosure does not replace independent confirmation of each of the five operations; there is no extra acknowledgment or billing input.

Requests use the fixed https://api.phantombuster.com/api/v2 origin. Reads make one GET to /agents/fetch, /agents/fetch-all or /agents/fetch-output; writes make one POST to /agents/launch or /agents/delete. List limit selects locally after validation, without a server query or pagination. No container resolution, raw log/event/result exposure, returned media, link following or polling is provided.

agent.launchRequest executes stored automation, potentially using saved platform credentials and causing scraping, external changes, quota use or charges. saveArgument:true can change saved launch defaults. agent.deleteRequest is destructive for the exact confirmed agentId; documented 400 missing-agent and 403 running-agent responses are failures. Each write makes one POST and returns only the LOCAL {requestedAgentId:<exact input>,httpStatus:200,outcomeVerified:false}. HTTP 200 does not prove launch, deletion, success or a container identity. The retained primary Swagger describes HTTP 200 without a write response-body schema; locally, an empty or bounded valid JSON body is fully scanned and discarded. Any failure after invocation leaves the remote outcome indeterminate. Do not automatically retry or read back; inspect Phantombuster separately before deciding what to do. Actual launch and delete were never performed during verification.

WORKFLOW_LIVE_PHANTOMBUSTER_AGENT_METADATA uses an explicit workspace boolean before the environment; otherwise exact false disables and unset enables. The live flag does not supply account or run authority. Saving remains local and unverified; actual connection testing and Review/provider/human verification are separate from synthetic local checks.

Each native operation has a 10-second budget starting before preparation, excluding outer database and independent confirmation latency; its gateway allows 11 seconds. Limits are 262144 decoded exposed response bytes, 8192 reader iterations including completion, JSON depth 16 and 10000 nodes. Overflow refuses rather than truncates. These are not bounds on raw wire bytes, decompression, allocation or operating-system execution. One attempt is used, with no automatic retry, readback, fallback, polling or returned-link fetch.

Full bounded upstream bodies, including discarded arguments, proxy passwords, storage, manifests or logs, may enter server memory before projection. Ordinary caller inputs, selected results, execution history and downstream data may be retained; this is not universal input redaction. Supported finite secret encodings are checked across the response, including ignored fields; arbitrary nested, mixed or encrypted encodings cannot all be recognized. Permissions, key acceptance, quota, account scope, human confirmation and real-provider behavior remain unverified.

See credential fields and block controls.

Pushcut and SIGNL4 account and attachment confirmation

Choose Pushcut with a saved workspace account of type pushcutApi, containing only {apiKey}, or SIGNL4 with signl4Api, containing only {teamSecret}. These are existing manually supplied secrets. Pushcut uses the raw API-Key header, not a secret notification-webhook URL key. SIGNL4 requires only the final opaque webhook team-secret segment, never the full URL or REST API key.

Save and reopen send zero provider HTTP and leave the account unverified. Test is unsupported, and no hidden option loader checks the account. Select the exact saved workspace account and independently confirm every operation, including metadata reads. Saving or accepting credential disclosure grants no execution authority. There is no extra acknowledgment or billing control. Direct workspace membership, organization membership and execute authority are required; organization-stored credentials are unsupported, and target-bound personal execution is untested.

For a SIGNL4 attachment, first produce a persisted file in the current workspace and pass its file descriptor under the selected Binary Property. Only PNG, JPG, JPEG, BMP, GIF, MP3 and WAV are supported, with a local 1 MiB limit checked before reading or copying. The file must still be ready, with matching size and SHA256 before and after the bounded read. Arbitrary bytes, URLs, local paths, filenames and MIME types are not accepted. Signature checks validate local magic/MIME, not full media decoding.

A standalone workflow pauses at that same SIGNL4 node after resolving the current file. Open its attachment confirmation, inspect the exact account ID, binary property, format, size and SHA256, then choose Approve attachment or Reject attachment. The approval expires after five minutes. Only the original actor can decide that exact request; changed accounts, changed files, stale decisions and replayed approvals are refused. The current file, account and execution permission are checked again after approval, together with live and cancellation state. The ordinary manual-run confirmation cannot authorize an upstream file whose digest is not yet known. Approval details omit secrets, private tokens and file content; normal file retention still applies.

Standalone placement includes the selected branch of a condition. Attachments in loop bodies, parallel branches, nested composite descendants, sub-workflows and single-step runs are unsupported and fail before preparation, approval callback or provider HTTP. No success marker or earlier approval bypasses this restriction; continuation for those placements remains future work. Ordinary Agent and human approvals keep their existing behavior.

Every operation makes one attempt, without automatic retries, fallback, readback, polling or returned-link fetching. Pushcut send returns an HTTP 200/202 receipt, with id and, for 202, scheduleTimestamp. SIGNL4 returns HTTP 201 with an optional nullable eventId; even {} is a valid selected result. Both envelopes keep outcomeVerified:false. A receipt proves neither delivery nor action execution, scheduling, alert creation, closure, acknowledgement or affected count. Resolve uses the original external correlation ID (externalId, sent as X-S4-ExternalID), never the returned eventId, and can affect multiple alerts. After invocation, a failure leaves the remote outcome indeterminate; inspect the provider separately before deciding whether to repeat a request.

Advanced execution boundaries: current account revisions use HMAC-bound request preparation and exact WeakMap input/context identity; neither a saved secret nor an approval marker creates authority. Mounted Approve/Reject uses the normal tRPC/REST approval route with the current original actor, exact signed confirmation view, canonical compare-and-set and replay protection. Only a valid signed scoped attachment wait is excluded from generic Human Task handling. Ordinary Agent and approval paths remain unchanged. These checks are local implementation properties, not real-provider or independent-human verification.

WORKFLOW_LIVE_PUSHCUT and WORKFLOW_LIVE_SIGNL4 use explicit workspace booleans before environment values; otherwise exact false disables and unset enables. Recheck current live/cancel state at dispatch. Organization-stored credentials are unsupported; target-bound personal execution remains untested.

The native 10-second budget starts before preparation and excludes outer database and independent-confirmation latency. Response limits are 256 KiB of decoded exposed bytes, 8192 reader iterations including completion, JSON depth 16 and 10000 nodes; overflow refuses rather than truncates. These are not raw-wire, decompression, allocation or operating-system bounds.

Full bounded private provider responses enter server memory before scanning and selected-output projection. Ordinary inputs, selected results, execution history, downstream data and workspace files may be retained; there is no universal input redaction. Finite known secret encodings are checked, without a guarantee for arbitrary mixed, nested or encrypted encodings. Provider permissions, routing, quota, costs, delivery, resolve affected count and actual human/provider behavior remain unverified. Pushcut configured actions can invoke shortcuts, web requests, online actions and other systems; dynamic content requires Pro, without an upgrade or billing-consent control. SIGNL4 routing can incur SMS, voice or escalation charges.

See all five operations and controls.

seven and MSG91 account selection and message confirmation

Save an existing seven key as sms77Api with exactly {apiKey}, or an existing MSG91 key as msg91Api with exactly {authkey}. Named password fields preserve exact text; Advanced JSON and Format JSON allow exactly that native pair within local 16 KiB/code-unit bounds. Keys use a local 1–256 visible ASCII grammar without spaces or trimming. Save/reopen send zero provider HTTP and stay unverified; Test, OAuth, hidden key loaders and template discovery are unavailable.

Select the exact current workspace account and independently confirm each send_sms, send_voice or send_template_sms invocation. Saving or reading a disclosure grants no run authority. Current direct workspace membership, organization membership, execute permission, account revision and prepared input identity are checked before dispatch, with current live and cancellation state. Organization-stored credentials are unsupported; target-bound personal execution is untested. There is no additional acknowledgment, billing or consent control.

seven exposes To text, Message textarea, optional SMS From text and an Options JSON-object textarea. SMS options preserve all six original keys: delay, foreign_id, flash, label, performance_tracking and ttl; voice options allows only from. Explicit false and permitted empty strings differ from absence. Whole-second timezone-bearing delay converts to Unix seconds, replacing original raw forwarding. Submitted contact/group entries can fan out; SMS encoding/segments and tracking affect costs, and tracking needs a configured custom domain. Voice uses a closed grammar without remote audio, DTMF callbacks, loops, call control or general SSML; existing account webhooks and recipient keypress events may still operate.

MSG91 explicitly migrates the original free-form sender/to/message action to an existing approved template_id and strict JSON recipient objects with exact case-sensitive variables. No automatic conversion or legacy fallback exists; current old-endpoint compatibility is unknown. Template Id and Short Url Expiry are text fields; Recipients is a textarea, not a comma fallback or multiselect. Short Url is a string0/1 select with a local string0 validation default; untouched config stays omitted. Real Time Response is an optional one-value string1 select without a default. Both selects retain Expression toggles; numeric1 is invalid. Expiry is absent unless supplied. The official example/CLI disagreement about optional fields remains unverified against accounts.

Each action makes one attempt, including with outer maxTries3, and no automatic retry, redirect, fallback, polling or readback. WORKFLOW_LIVE_SEVEN and WORKFLOW_LIVE_MSG91 prefer explicit workspace booleans; otherwise exact environment false disables and unset enables. seven string100 is acceptance only and101 may be partial; MSG91 success returns a selected request ID. All receipts retain outcomeVerified:false without a delivery, price or template-approval guarantee. Post-invocation failures are uncertain; inspect the provider separately before deciding to repeat.

The native10s budget starts before preparation and excludes outer database and independent human-confirmation latency. Responses are bounded to 256 KiB decoded exposed bytes,8192 reads including completion,depth16 and10000 JSON nodes, not raw-wire, decompression, allocation or OS bounds. Full bounded responses enter memory before projection; ordinary inputs, selected outputs, history and downstream data may be retained. Known-secret scanning is not universal redaction. Message data goes to providers and recipients and may incur costs. Real acceptance, account rights, costs, delivery, Review and independent human confirmation remain unverified; local/synthetic proof does not resolve retained REA unknowns or establish n8n parity.

See credential JSON and exact fields, defaults and outputs.

YOURLS installation credentials and public QuickChart

YOURLS uses an existing yourlsApi workspace account with exactly two native fields:

{ "signature": "YOUR_EXISTING_SIGNATURE", "url": "https://links.example.org/tools" }

Enter the signature in Existing YOURLS signature (unverified) and the installation base in Public HTTPS installation URL. Include an installation subdirectory when needed; omit /yourls-api.php from this field. The installation must use public DNS and HTTPS port 443. Direct API paths, HTTP/private installations, literal IPs, query strings, fragments and URL credentials are refused. There is no OAuth, automatic installation discovery or guarantee for custom plugins or authentication. Exact text is preserved; do not rely on trimming.

Named fields, Advanced: edit JSON and Format JSON only edit local credential material. Save and reopen send zero provider HTTP requests, remain unverified and offer no Test. Choose the exact current workspace account before running. Save does not grant execution authority: current direct workspace and organization membership, execute permission, account revision, HMAC material and prepared-object identity are checked for dispatch, along with live/cancel state. Organization-stored credentials are unsupported. WORKFLOW_LIVE_YOURLS controls this provider through the existing workspace/environment live policy.

QuickChart is public: authentication:none, credentials:[]. Do not create a fake API key, manual account or Connect account entry. Its public card says Public render · no account required and Independent confirmation for every render. Public access still requires current workspace and organization authority and a signed, independently confirmed render. WORKFLOW_LIVE_QUICKCHART controls execution. Saving a YOURLS account is separate from consenting to any YOURLS action or public chart disclosure.

Each shorten, expand, stats and render invocation requires independent confirmation and makes one attempt even with outer maxTries3. There is no automatic retry, redirect, fallback, polling or readback. Full bounded responses enter memory before selected projection; ordinary inputs, selected outputs, run history, downstream data and chart files may be retained. Known secret encodings are checked, without a claim to detect arbitrary encodings. A failure after invocation leaves the provider or local file outcome uncertain; inspect independently before deciding to repeat. Local PostgreSQL tests cover current workspace/organization authority and revocation using synthetic guarded HTTP responses. QuickChart file storage in those tests is an in-memory fixture. Real provider requests, DNS/TLS, MinIO/S3, ClamAV, independent human approval and hosted Review remain unverified.

YOURLS and QuickChart finite operations

The native YOURLS provider exposes shorten / yourls_shorten, expand / yourls_expand and stats / yourls_stats. QuickChart exposes only render / quickchart_render, with no credentials. All four operations have required effective confirmation. They have no import aliases or automatically materialized Tool wrappers; native n8n item-loop, raw-output and continueOnFail parity are not claimed.

YOURLS sends a fresh POST form to the validated installation/subdirectory's yourls-api.php. The signature stays out of the URL. Shorten accepts required url and optional additionalFields with only keyword and title strings:

{ "url": "https://example.org/report", "additionalFields": { "keyword": "report", "title": "Quarterly report" } }

Omitted additional fields differ from explicit empty strings; an omitted title may cause YOURLS to fetch the target title using its own network. Shorten returns accepted:true,outcomeVerified:false,shorturl with optional selected url metadata; this can reuse an existing link and never proves new persistence. Expand and stats accept exactly {"shortUrl":"https://links.example.org/report"}. Expand returns required longurl and optional keyword/shorturl/title. Stats returns selected shorturl/url/title/timestamp/clicks; clicks is an integer lexeme string. IP/plugin fields are omitted. Returned URLs are metadata and are never followed.

Public transport requires every A/AAAA address to be public on two matching resolutions. Approved addresses are bound to the TLS connection with the original hostname, SNI and certificate verification, without environment proxy/allowlist bypasses, redirects or retries. Synthetic DNS and socket dependencies test local decisions; they do not prove real DNS/TLS. Database authority checks and socket initiation are not atomic.

QuickChart sends only fixed POST https://quickchart.io/chart, pins Chart.js major 2 and requires actual PNG, PDF, SVG or WebP bytes rather than a URL substitute. Its input is an explicit normalized items batch, not an adapter that runs original n8n items. First-item labels/options/output become the top-level fields; the last effective item type controls the chart type. Each dataset's chartOptions.horizontal overrides the top-level first-options fallback. Supported types are bar, doughnut, line, pie and polarArea.

{
  "items": [
    { "chartType": "bar", "data": [10, 20], "datasetOptions": { "label": "Sales", "backgroundColor": "#336699" } }
  ],
  "labelsMode": "array",
  "labelsArray": ["Q1", "Q2"],
  "output": "data",
  "chartOptions": { "width": 500, "height": 300, "devicePixelRatio": 2, "format": "png", "horizontal": false }
}

There are 1–50 datasets and 1–250 points per dataset. Labels support the manual labelsUi object or a direct array/strict JSON text; numeric JSON text must preserve its exact lexeme. Dataset options allow label, backgroundColor, borderColor, fill and pointStyle. Chart options allow backgroundColor, ratio1/2, dimensions1–2000, format and horizontal. Defaults are bar, manually, output data, 500×300, ratio2, PNG and horizontal false; dataset labels fall back to Chart. Declared JSON only: no JavaScript functions, plugins, remote resources or arbitrary rich configuration. These local limits are separate from vendor quotas, cost and rendering.

The output has json.chart, binary[output] and outcomeVerified:false. Each binary entry contains reference, mimeType, format, size and sha256; it represents actual bounded bytes. The trusted write-only WorkspaceFileService path creates/uploads/completes/scans/adds a typed run reference and compensates owned files/references on failure. Exact admitted run IDs include CUIDs. Current workspace AND organization, execute, actor, live, run ownership, workflow identity and cancellation are checked around asynchronous work. One file is bounded to1 MiB; an aggregate workspace quota is not available or claimed. Files are attachments, with ordinary file/history/downstream retention and scanner rules.

The native10s budget starts before preparation, excluding outer database and human-confirmation latency. JSON is bounded to256 KiB, depth16/10000 nodes and8192 reads; images have a separate 1 MiB cap and raster4M-pixel limit. Restricted local PNG/WebP decoding and SVG/PDF profiles are not full viewer/renderer conformance or wire/OS/allocation guarantees. Full bodies enter memory. Local PostgreSQL tests cover current workspace/organization authority and revocation with synthetic guarded HTTP responses; QuickChart attachment storage remains an in-memory fixture in those tests. Real provider requests, DNS/TLS, MinIO/S3, ClamAV, human approval, AI request phrasing through persistence and hosted Review remain unverified.

Vero and Vonage: choose an account and review each action

Save an existing Vero Track authToken as veroApi, or Vonage apiKey and apiSecret as vonageApi, in the workspace you will run from. Named fields, Advanced JSON and Format JSON use strict local validation. Save makes zero HTTP calls, remains unverified and grants no run authority; Test is unsupported. Neither provider adds OAuth or new acknowledgment fields. Select the exact current account instead of relying on its display name or another account. Organization-stored credentials are unsupported.

Vero supports create/update, alias, unsubscribe, resubscribe, delete, addTags, removeTags and track. Alias may merge users, deletion is irreversible, subscriptions are global, and events may create users or trigger campaigns. Choose carefully before confirming. Json Parameters selects UI key/value JSON objects or JSON text; nested null, false, empty values and exact numeric text remain meaningful. Tags retain spaces and empty comma-separated entries.

Vonage sends text, Unicode or literal binary hex plus UDH, not a workspace file. Additional Fields has seven original descriptors, six executable: callback, client-ref, message-class, protocol-id, status-report-req and TTL minutes. False delivery reports, class zero and empty client-ref are intentional. Legacy account-ref, WAP Push, vCard/vCal and unused binaryPropertyName refuse before HTTP without a text fallback. See exact field controls, defaults and JSON examples.

These are explicit migrations: Vero uses JSON objects and query auth_token instead of its ambiguous original body/form helper; Vonage uses fixed Basic-auth SMS instead of original form credentials. There is no legacy fallback. Every action independently confirms the exact input/account, including any Vonage callback. A callback asks Vonage to contact an external HTTPS endpoint; ownership, vendor DNS, sockets, redirects and delivery are not controlled or proved. Confluye adds no receiver, polling or receipt proof. SMS may incur segment charges.

Current principal, direct workspace AND current organization membership, execute permission, selected account revision, live setting, workflow/run and cancellation remain required around awaited work, including the transport check after DNS before HTTP. Database and socket state are not atomic. One attempt occurs even with maxTries3. outcomeVerified:false is an unverified receipt; a lost or nonzero response leaves uncertain effects, so investigate before repeating. Ordinary inputs, selected outputs, history and downstream retention remain. Local authority tests used PostgreSQL with synthetic HTTP/confirmation; live APIs, delivery, human consent, Review and Vero/Vonage production-browser behavior remain unverified.

UpLead and OneSimpleAPI: existing accounts and confirmed utilities

UpLead enriches company and person records. OneSimpleAPI provides ten utility definitions: nine support requests and Instagram is explicitly unavailable. These are selected native integrations, not full n8n node, item-loop, credential-import, error-mode or live-provider parity. The original ordinary and Tool resource/operation aliases identify these definitions; they do not import saved credentials or certify all original options.

Choose a manually acquired provider account, then save and select the exact account for the workflow workspace. UpLead uses upleadApi with only apiKey, sent as raw Authorization, never an added Bearer prefix. OneSimpleAPI uses oneSimpleApi with only apiToken, sent as query token with output=json. A provider's individually acquired account is distinct from the application's personal storage scope. Personal accounts can execute through an exact saved-account binding owned by the current actor, with current workspace and organization permissions; real-provider and Review behavior remains unverified. Organization-stored credentials are unsupported. An organization workspace still requires current workspace and organization authority.

Named password fields, Advanced: edit JSON and Format JSON validate locally. Save sends zero provider requests, leaves authentication unverified and grants no execution authority. Test is unavailable; there is no OAuth flow, imported credential-name fallback, extra acknowledgment checkbox or billing control. Never paste account secrets into operation fields or URLs.

Independently confirm every invocation, including metered GET lookups. Review the exact operation, selected account and complete input. Current actor, workspace/organization membership, execute permission, account revision, live preference, run/workflow/version and cancellation are checked again after awaited work and at the guarded transport boundary after provider DNS. These checks do not create an atomic database-to-socket guarantee. Each invocation makes at most one attempt even when workflow maxTries=3 requests retries. A failure after invocation leaves an uncertain outcome; inspect it before manually repeating. outcomeVerified:false means there is no independent verification of the effect, not a retry instruction or a success certificate.

Inputs and selected results can remain in ordinary run history and downstream steps. Full bounded responses, including later-discarded fields, are received and scanned for known secret material before projection; this is not universal secret detection. JSON numeric lexemes are preserved as decimal strings, avoiding JavaScript rounding. Local 10-second/256-KiB and parser/read limits do not guarantee vendor, wire, allocation or operating-system bounds. Account validity, quotas, costs, current provider contracts, hosted Review flows and human confirmation remain unverified.

Save and reopen examples

Use the named password field or Advanced JSON with exactly {"apiKey":"YOUR_EXISTING_KEY"} for UpLead, or {"apiToken":"YOUR_EXISTING_TOKEN"} for OneSimpleAPI. Format only formats valid JSON locally. Save/reopen keeps the chosen account; select it explicitly for the workflow. These examples are placeholders, not operational credentials. Do not interpret an unverified saved account as a tested connection. A separate account is needed only when you intend to use that provider, not to view these reference pages.

UpLead input and result choices

company.enrich uses uplead_company_enrich: provide domain or a nonblank company. person.enrich uses uplead_person_enrich: provide email, or nonblank firstname, lastname and domain. Field names keep the original casing. Original resource defaults are company and enrich; the app selects an explicit operation. All identifier controls start as empty text. Original descriptors did not mark these fields required, but the native validator requires one complete lookup alternative. Whitespace is preserved, not silently trimmed into a different lookup; domain/email syntax and bounded strings are validated. An empty optional value stays meaningful, while a wholly empty lookup is rejected before HTTP.

The fixed GET destinations are https://api.uplead.com/v2/company-search and /v2/person-search; identifiers are shared with UpLead and may consume credits. Selected company fields include identity/domain, address/contact data, industry/classification, revenue/employees and social URL metadata; selected person fields include identity, job/contact data, location, employer/domain and social URL metadata. Available-credit data and unknown fields are omitted after scanning the full bounded response. Output is data plus outcomeVerified:false; data may be an object, array or null. A null/no-match result does not prove nonexistence, and optional field presence is not guaranteed. Returned URL fields are never fetched.

OneSimpleAPI operation choices

The original resource defaults are website/pdf, socialProfile/instagramProfile, information/exchangeRate and utility/validateEmail; the app uses explicit definitions. Native credential auth remains query token plus output=json, separate from the input field named output.

Options is one JSON object textarea, not separate nested widgets. A selected boolean stays boolean, including false; omitted options remain omitted. PDF's original selected-option defaults are page empty/force false; screenshot's are screen empty/force false/fullpage false; SEO's headers false. These describe original option selection, not insertion of every option into an empty object. Unknown keys and malformed/type-mismatched values are rejected. Original output was shown only for download:true; the native descriptor exposes it with default data, while binary mode still refuses. It must be an identifier-shaped string and does not rename the native data result envelope.

Text inputs support expressions that retain the resolved type; defaults do not silently stringify an object or number. The download switch is a literal boolean. Operation controls use the original labels and actual native field names; the Help controls and disclosures are translated, but this is not a claim that every raw operation label is localized.

URL disclosure and unavailable modes

PDF, screenshot, SEO, image metadata and URL expansion ask OneSimpleAPI to retrieve the exact approved public HTTPS link. The local syntax policy excludes credentials, fragments, IP literals, private/local names, non-HTTPS and nonstandard ports. Confirmation approves the exact input and current account/run authority before transport; the application does not probe the requested URL, establish target ownership, or control the vendor's DNS, sockets, redirects or remote rendering. No remote-safety guarantee follows from local URL validation. Returned links are metadata and are never followed or downloaded here.

PDF, screenshot and QR require download:false. download:true refuses before HTTP: there is no link substituted as a successful binary download and no workspace file. Instagram also refuses before HTTP because its current contract is unknown; a cached documentation 404 does not prove removal. Original QR Small/Medium/Large and PNG/SVG casing and image raw=true are retained but current vendor acceptance/semantics remain unknown. Screenshot fullpage explicitly maps to documented full=yes/no. Only the listed original options are accepted; current vendor features such as PDF background/HTML, QR colors or other output formats are not added implicitly.

Use LoneScale lists and public PostBin bins

LoneScale uses native lonescaleApi (lowercase s) with exactly apiKey. Explicitly resave original loneScaleApi; there is no generic alias or fallback. Requests send x-api-key only to public-api.lonescale.com. The original overlapping Authorization/X-API-KEY intent is replaced explicitly: current endpoint security schemas specify x-api-key, although overview prose mentions Authorization. Save, reopen, named fields, Advanced JSON and Format send zero provider HTTP and leave the account unverified. Test, OAuth, credential discovery and key generation are unavailable. Select the exact saved account for the workflow workspace.

PostBin has authentication none and credentials:[]. Do not create or select a saved account, key, credential form, Test connection or OAuth flow. Its public bin/request IDs establish neither ownership nor privacy; use only bins and data you are authorized to access.

createList takes required name and Type (COMPANY by default, or PEOPLE). It sends name and entity. addItem defaults to PEOPLE and requires a list UUID. Original people mode requires first_name and last_name, plus optional peopleAdditionalFields: full_name, email, company_name, current_position, domain, linkedin_url, location, contact_id. Original company mode uses company_name, separate native name, and optional companyAdditionalFields: linkedin_url, domain, location, contact_id.

Alternatively supply fields JSON alone, without original top-level identity fields or either additional-fields collection. Supported keys are first_name, last_name, full_name, linkedin_url, company_name, current_position, domain, location, name, email, contact_id. PEOPLE requires two nonempty documented identity fields from linkedin_url, name, location, first_name, last_name, full_name, company_name, current_position; email/domain/contact_id alone do not satisfy this rule. COMPANY requires domain OR separate name and location. company_name never silently becomes name. Entity-incompatible fields and mixed input modes refuse before HTTP. Strings preserve case, spaces and numeric-looking values such as "0007"; empty optional strings follow original truthy-string omission. Email and LinkedIn URL syntax is validated; this does not prove identity.

listOptions replaces the automatic list loader with one explicit confirmed GET: choose Type and zero-based page (default 0, maximum 1000000). The documented page contains ten records; selected options filter entity and preserve hasNextPage, total and incompleteness. Supply the next page explicitly; there is no automatic pagination or Save/Test request. Workflow options, workflow-hook create/check/delete and LoneScale triggers remain unavailable because current route/authentication/output, callback authenticity and durable ownership contracts are unresolved. GET /webhooks is a different contract and is never substituted.

PostBin supports createBin, getBin, deleteBin, getRequest, removeFirstRequest and sendRequest. Bin creation has no required content. Get/delete require binId; get request also requires requestId; consume-first requires binId; send requires binId and defaults binContent to an empty string. Use a finite safe opaque ID (ASCII letters, digits, _ or -, at most128 characters, beginning with a letter or digit), or an exact supported original numeric bin URL/label. This explicit migration accepts documented opaque IDs as well as the original numeric form. Bin IDs api/b and request ID shift are refused where they would select another route. A read alias never consumes the first request.

removeFirstRequest uses GET /api/bin/{binId}/req/shift but removes data: treat it as an external write and confirm it independently. deleteBin uses DELETE; HTTP200 does not prove that the bin previously existed. sendRequest POSTs native JSON {"content":string} to the exact bin: this is an explicit serializer migration, not proof of the original generic serializer. Empty text is intentional. Stored false, null, zero, empty strings/arrays/objects remain distinct; selected JSON number lexemes become exact strings without rounding. Bin results expose selected ID, numeric timestamps/ISO forms and constructed request/view URLs. Request results expose selected method/path/headers/query/body/IP/timestamps; send returns a request ID. Returned links are never followed and no readback proves delivery.

Request Options is one JSON-object textarea, not separate native switches. All four original options have an explicit disposition:

The actual HOST item plan validates every item before the first HTTP request, accepts at most100 items with the same pacing policy, and executes sequentially. Aggregate interbatch waiting is at most30000 ms. It confirms and rechecks current authority for each item, pairs each result with its input and stops after failure while retaining prior receipts in failed output. It does not automatically resume or provide durable continuation. The custom timeout starts with each native invocation; outer pacing, confirmation and database latency are separate. Local limits do not prove wire/allocation or operating-system confinement.

Every action, including reads and list options, needs independent confirmation of its exact input. Each action has one attempt even with imported maxTries:3. Successful and handled-failure receipts keep outcomeVerified:false; a failed write after invocation may already have changed the provider. Investigate uncertainty before repeating. Error routing does not turn failure into success, retry or compensation. Inputs, selected outputs, run history and downstream data use ordinary retention. Stored PostBin headers, body and IP can reveal personal data.

Current actor, workspace AND organization membership, execute permission, selected account revision where applicable, live run and cancellation are checked again around asynchronous work. This does not provide an atomic database/socket guarantee. Workspace accounts and explicitly selected personal native accounts resolve by exact credential ID and credential family. A personal account must belong to the authenticated run actor in the workflow workspace, with current workspace AND organization permission; its display name grants no authority and cannot select another account. Organization-stored credentials are unsupported. Missing or changed persisted run actors refuse before dispatch. A personally acquired provider key is distinct from its application storage scope. Local and synthetic tests do not prove real provider acceptance, organization persistence, independent human consent or Review behavior. No full n8n parity or durable continuation is claimed.

Stackby, Dropcontact and Oura accounts

Save an existing Stackby key as stackbyApi or a Dropcontact key as dropcontactApi, with exactly {"apiKey":"YOUR_EXISTING_KEY"}. Stackby uses the API Beta profile at https://stackby.com/api/betav1/ with api-key; other Stackby profiles are separate contracts and are not tried as fallbacks. Dropcontact uses https://api.dropcontact.com/v1/enrich/all with X-Access-Token; original .io/batch requests require this explicit migration.

Oura uses the new ouraOAuthAccessToken family with exactly {"authType":"oauth_access_token","accessToken":"YOUR_EXISTING_OAUTH_ACCESS_TOKEN"}. Supply an existing OAuth access token yourself. Retired Personal Access Tokens and original ouraApi credentials are unsupported; resaving opaque bytes does not establish token type, expiry or consent. Profile reads require personal; daily scores require daily. These are required grants, not verified consent or scope metadata. OAuth acquisition, refresh, introspection and callbacks are unavailable. Stackby and Dropcontact retain only their selected ordinary and Tool operation aliases. Oura is a separate manual-token counterpart with no n8n import aliases, including Tool aliases; migrate its inputs and account explicitly.

Use named fields or Advanced JSON; Format validates and formats locally. Save and reopen send zero provider requests and leave authentication unverified. Test connection is unavailable. Select the exact saved account for this workflow: workspace accounts or a personal account owned by the current actor can be used. Current workspace and organization permissions still apply; organization-stored accounts are unsupported. A queued run needs its original persisted actor, and changing an account, actor, workflow version or permission can invalidate an earlier approval.

Confirm the exact input independently for every physical request, including reads, later pages and status checks. Each request has one attempt even with imported maxTries:3. outcomeVerified:false remains explicit after success or a handled failure; a failed write may already have changed the provider. Investigate uncertainty before repeating. Inputs, selected results, ordinary history and downstream steps may retain data. Local and simulated checks do not establish real provider compatibility, token grants or an authenticated human decision.

Stackby rows and paging

Choose row.append, row.delete, row.list or row.read. Supply a Stack ID and table name. Read also takes a record ID; delete takes an array of one to ten row IDs. Append accepts one to ten records, each with a field object, or Original items with each item's Stack ID, table, comma-separated columns and item JSON. Original-item migration preserves destination grouping and input indices, splits physical writes into groups of at most ten rows, and retains every batch receipt. A partial failure does not erase earlier receipts or authorize automatic replay. Read, list and append require outputFields, an array of one to fifty field names; missing vendor values remain missing, separately from false, null and empty values.

List takes an explicit Page or Complete mode. Its original returnAll:true and limit:1000 defaults express intent; they do not establish completeness. Native controls include offset (default 0, maximum 1000000), page size (default 100, range 1..100), limit (1..1000) and view (default empty text). Page mode returns an explicitly incomplete page. Complete migration persists its plan and waits between windows; every later physical page needs a fresh independent confirmation. It stops on an observed empty page or a local bound, without inventing a total or guaranteeing all provider rows. Process-local pacing cannot control other clients. API Beta compatibility remains unverified with a real account.

Dropcontact enrichment and status

Choose enrich to submit a contact or fetchRequest to read one exact requestId. Enrichment has an Email field (default empty text), Additional Fields JSON (default {}) and Options JSON (default {}). Additional fields are num_siren, siret, company, country, first_name, full_name, last_name, linkedin, phone and website. Options accept waitTime in seconds (1..86400, effective default 45), siren and language (en or fr). Original defaults are Simplify false, French company enrichment false and language English. Preserve explicit false and omitted options separately.

Enable Simplify (submit job only) for a single submission. Its request ID does not establish completed enrichment. With the original Simplify false default, the workflow persists a configurable timer (45 seconds by default), then opens a separate Human Task before one confirmed status GET. A pending result stays pending; there is no hidden polling. Completion migration needs the durable runtime and Human Tasks and refuses before sending inside loops, parallel blocks, subworkflows and single-step execution. Submit-only and explicit status reads are separate available choices.

A custom public HTTPS callback URL asks Dropcontact to deliver contact data to that exact address. Existing account callbacks may also receive it. Review this sharing before confirmation; the application neither probes the URL nor establishes receiver ownership, vendor DNS, redirects or delivery. Contact requests may consume credits. The ten-second physical-request budget excludes the outer timer, human decision and database waits.

Oura profile and daily scores

Choose profile.get, summary.getActivity, summary.getReadiness or summary.getSleep. Profile has no input fields and excludes email from its selected result. Sleep returns a daily sleep score, rather than sleep periods. Profile and summary numeric JSON values are returned as exact decimal text; omitted optional fields remain omitted and null remains null.

Every summary requires explicit startDate and endDate calendar dates in YYYY-MM-DD, with at most 31 inclusive days. This replaces the original timezone-dependent datetime and omitted-date defaults. Limit is 1..10 (default 5), applied locally and never sent to Oura. Optional nextToken selects one explicit provider page; there is no Return All or automatic pagination. receivedCount, returnedCount, omittedCount and locallyTruncated describe only that page. upstreamHasNextPage and nextToken describe upstream continuation: advancing the token cannot recover rows omitted locally from the preceding page. Raising the limit and rerunning requires another confirmation and is not a completeness guarantee.

Oura receives personal and health-data requests. The full bounded upstream response, including fields later discarded, enters memory before projection; selected outputs and ordinary history/downstream retention remain. No live token, identity, subscription, consent, scope or health-data accuracy verification is implied. Flow remains research-only because authority for its former API could not be established; it is not an available integration.

Zammad, Wekan, Quickbase and Salesmate accounts

Select the exact saved account and matching credential family in the workflow. Workspace accounts and personal accounts owned by the current actor are eligible; organization-stored accounts are unsupported. Current workspace and organization execution permissions still apply. Saving does not authorize a request or verify token identity, provider rights or consent. Named fields, Advanced JSON and Format validate the strict local shape; unknown, duplicate or invalid fields are refused without silently repairing the input. Save and reopen send no provider request and leave authentication unverified. Test connection, automatic login and OAuth acquisition or refresh are unavailable for these profiles.

Zammad: explicit Basic or Token mode

Use zammadNativeBasicApi for Basic operations:

{
  "mode": "basicAuth",
  "baseUrl": "https://support.example.com",
  "username": "YOUR_EXISTING_USERNAME",
  "password": "YOUR_EXISTING_PASSWORD"
}

Use zammadNativeTokenApi for Token operations:

{ "mode": "tokenAuth", "baseUrl": "https://support.example.com", "accessToken": "YOUR_EXISTING_TOKEN" }

Choose the corresponding basicAuth.<action> or tokenAuth.<action> operation: 25 logical actions become 50 mode-qualified choices. Use the public HTTPS installation root or supported path prefix. Token mode uses Zammad's Token token= authorization. OAuth Bearer and original zammadBasicAuthApi or zammadTokenAuthApi credentials require explicit migration; no family or server fallback is tried. No n8n import aliases are declared.

Use names where a field asks for names: organization and decimal organizationId are mutually exclusive. Create needs group/organization name, user firstname and lastname, or ticket title, group, customer and article. Nested fields JSON must follow the selected resource shape; empty updates and unsafe or duplicate custom field names are refused. Supply the selected installation's custom-field names manually.

Lists, user/field options and article lists request one explicit page: page defaults to 1, perPage to 50, maximum 100. Every later page or option read needs a separate confirmation; nextPageCandidate is unverified and complete:false does not promise an exhaustive result. Migrate Return All or larger imported limits manually. Ticket creation returns its POST response without an article readback; confirm article.list separately. Article upload and attachment download use actual owned workspace files, at most 256 KiB, with current ownership, scope, ready scan state and integrity checks. Download requires ticket/article/attachment IDs, file name and MIME type. A URL is not a file and is never followed implicitly.

Wekan: existing Bearer token and selected user

Save wekanExistingBearerApi with exactly:

{
  "accountOrigin": "https://boards.example.com",
  "token": "YOUR_EXISTING_BEARER_TOKEN",
  "userId": "YOUR_EXISTING_USER_ID"
}

Use the public HTTPS installation origin or supported path prefix and supply the original opaque userId yourself. Declaring a user ID does not prove token ownership. Original wekanApi username/password credentials need explicit migration. The selected static contract targets Wekan v12.57; installed-server capability and permissions remain unverified.

There are 35 choices: 24 ordinary actions, three explicitly named runtimeChecklist actions and eight independent options reads. Only the 24 ordinary actions declare ordinary Wekan aliases; Tool aliases, triggers and attachment operations are unavailable. Each options read needs separate confirmation, and listing users may require server administrator rights.

Board owner and card/comment create author must match the selected user. Board role overrides and later card-author changes are refused. Use opaque IDs, not routes or encoded separators. For card.getAll, provide the list or swimlane ID selected by fromObject. Collections return one response, locally limited to 1–200 records, default 100; returnAll still means this bounded response, completeness is unknown and nextPage is null. There is no cursor or hidden pagination. Explicit false, zero, empty text and omission remain distinct locally. Card dates support null/empty clearing and members/assignees support empty arrays, but the server may ignore falsy updates, filter inactive members or truncate an updated card title to 1,000 characters.

Quickbase: US realm and explicit field IDs

Save quickbaseApi with exactly:

{ "authType": "user_token", "hostname": "your-realm.quickbase.com", "userToken": "YOUR_EXISTING_USER_TOKEN" }

hostname is the exact US realm hostname, without scheme or path. Requests use fixed https://api.quickbase.com/v1; EU/custom origins are unavailable. Resave original QuickBaseApi credentials explicitly. Ten ordinary actions retain ordinary and Tool aliases; getTableFields and getUniqueTableFields are two separate first-class discovery actions without aliases.

Supply table and field IDs manually. Runtime requirements still apply even when the shared form marks a field optional: a table request needs tableId, and record writes need the selected fields/items. Record writes require options.useFieldIDs:true; Update also requires simple:false and explicit Record ID field 3 matching updateKey. Hidden field discovery followed by a write and multiple physical request units are unavailable. Discover fields in a separate approved action, then configure IDs. Create, Update and Upsert share the vendor insert/update endpoint and may insert or update.

A batch has 1–10 original items, sharing item 0 table/options and per-item columns; in explicit-ID mode every item key is sent, regardless of columns. Keep itemParameters and itemBinary cardinality/order aligned. Missing or ambiguous labels/IDs and collisions refuse rather than inventing values. Each physical request, including reads and discovery, needs its own signed exact-request Human Task approval by the original actor. Review method, URL, realm, query, complete body and input indexes in the preview, at most 48 KiB. The initial path permits one physical request. Generic Run or single-step acknowledgment cannot replace this approval.

Reads return one bounded page, up to 100 records; Return All is still one page. Advance skip manually after reviewing incomplete/stalled/nextCandidate metadata; there is no snapshot or exhaustive-result guarantee. Raw envelopes remain available and JSON numeric lexemes stay exact strings; HTTP 207 line errors can mean a partial result. Downloads decode at most 1 MiB of actual bytes into owned workspace files with scan, size and SHA256 checks; returned URLs are not fetched as files.

Salesmate: existing v4 tenant token and flat CRUD fields

Save salesmateV4AccessToken with exactly:

{
  "profile": "salesmate-v4-access-token",
  "tenantOrigin": "https://your-tenant.salesmate.io",
  "accessToken": "YOUR_EXISTING_V4_ACCESS_TOKEN"
}

Use the exact lowercase HTTPS tenant origin without trailing slash, path, port, credentials, query or fragment. Central/reserved hosts, legacy salesmateApi/sessionToken and v1/v2 endpoints require explicit migration; no fallback is tried. The durable runtime with persisted authority is required; memory-only execution cannot authorize these requests.

Twelve actions create, get, update or delete one activity, company or deal. No n8n import aliases are declared. GetAll/search, custom fields, activity dueDate, bulk IDs, hidden loaders and the old Get projection are unavailable. Get requires rawData:true; create/update default to rawData:false. Get/update/delete IDs are positive decimal text, up to 20 digits. owner, primaryContact and primaryCompany use the declared numeric controls. A provider owner ID grants no app authority.

Updates require all mandatory creation fields: activity title, owner, type; company name, owner; deal title, owner, primaryContact, current pipeline/stage names, status (Open, Close or Lost) and uppercase three-letter currency. Supply pipeline/stage names manually. Migrate the original Billing State/Country collision into separate billingState and billingCountry; overwritten values cannot be reconstructed. Explicit false and zero remain supported where declared.

Author dealValue as decimal text, including expressions returning text: at most 18 total digits and nine fractional digits, optional minus sign, no plus or exponent. Preserve digits, scale, zero and negative zero. Empty text is invalid; omission remains omitted. Migrate imported numeric workflow values explicitly; already-rounded IEEE754 digits cannot be recovered. For example, use the string "123456789.123456789" without converting through Number.

Every action, including Get, requires its own durable exact-request Human Task decision by the original persisted actor. Generic Run approval and provider owner IDs are insufficient. Input preview is bounded to 24 KiB, the signed payload to 32 KiB, and approval expires after 15 minutes. Account, configuration or authority changes can invalidate it sooner. CRM writes may invite/notify people, trigger account automations and consume quota or credits. Delete is destructive; the activity soft-delete request choice does not guarantee recovery. Returned provider JSON retains numeric lexemes as strings; links remain data, without file transfer or implicit fetches.

Confirm each request and inspect uncertain effects

All four profiles require independent confirmation and current account, original actor, workspace and organization authority, workflow/configuration revision and live-run/cancellation checks. Each invocation has one attempt even when workflow retries are enabled. No automatic retry, polling, readback or compensation verifies a change. outcomeVerified:false leaves effects unverified after HTTP success; inspect uncertain changes before intentionally repeating them. Inputs, full bounded provider data, selected results and ordinary history/downstream data may persist. Saving an account or accepting a confirmation does not prove real-provider compatibility, permissions, scanning or delivery. These selected counterparts do not establish complete n8n action, field, authentication, import or runtime parity.

MISP accounts and exact requests

Save an existing MISP account as mispApi with exactly:

{ "apiBaseUrl": "https://misp.example.org/security/misp", "apiKey": "YOUR_EXISTING_RAW_API_KEY" }

Use the public HTTPS installation URL, including its exact path prefix, and an existing raw API key. Save and reopen are local and leave authentication unverified. Test connection, OAuth and automatic key acquisition are unavailable. Select the exact workspace account or a personal account owned by the run actor. Current workspace and organization execution permissions apply; organization-stored accounts are unsupported.

The 44 choices cover attributes, events, event tags, feeds, galaxies, notice lists, objects, organisations, tags, users and warning lists. Ordinary MISP and MISP Tool imports identify these covered operations; they do not establish general n8n Tool runtime compatibility. Supply IDs manually as positive decimal text with at most 18 digits. UUIDs and larger IDs require an explicit migration. Additional Fields and Update Fields accept only the declared keys; updates must contain a supported change.

For attribute, event and object search, leave Use Json off and fill Value plus optional Additional Fields. To use Json Output, turn Use Json on and clear Value and Additional Fields. JSON accepts the finite search fields value, category, deleted, searchall, tags, type, published, page, limit and returnFormat:"json"; it is not an arbitrary request body. Explicit false values remain false. GetAll returns one bounded response, with a local limit of 1–1,000 (default 50). Return All does not guarantee completeness or request another page. There are no hidden option lookups, retries or readbacks.

The original user.invited_by, organisation.usercount, organisation.created_by_email and feed-creation additionalFields.json effects are unsupported and refused. Feed update rules use their declared finite JSON shape. Insecure TLS, binary workflows and triggers are unavailable. Original source versions and the selected current API contract differ; installed-instance permissions and provider compatibility remain unverified.

Every read and write pauses for a separate exact-request Human Task assigned to the original run actor. Review the selected account, method, URL, complete body and described effect. Approval expires after 15 minutes and can become invalid sooner if the account, workflow, run, input or permissions change. The preview is limited to 24 KiB and its signed payload to 32 KiB. Another Member or Owner cannot approve on the original actor's behalf. Durable workflow execution and Human Tasks must be enabled.

Publishing an event may queue publication and fan out to other systems. Creating a user may email credentials and create an API key. Deletion can remove provider data. Each confirmed request has one attempt even when workflow retries are enabled. outcomeVerified:false remains explicit: a receipt does not independently verify the change, queued publication or delivery. After an uncertain failure, inspect the provider before intentionally repeating the request. Inputs and selected provider data may remain in run history and downstream steps; credential-like response fields are excluded, but ordinary content can still contain sensitive data.

SeaTable accounts and staged requests

Save an existing account as seaTableApi. For example:

{
  "profile": "seatable-staged-v2",
  "kind": "api-token",
  "environment": "selfHosted",
  "domain": "https://tables.example.org/team",
  "baseUuid": "11111111-1111-4111-8111-111111111111",
  "providerWorkspaceId": 42,
  "permission": "write",
  "token": "YOUR_EXISTING_TOKEN"
}

Choose cloudHosted with exactly https://cloud.seatable.io, or selfHosted with the exact public HTTPS installation and optional subpath. Enter the selected base UUID, SeaTable workspace ID and read or write permission explicitly. Save/reopen validate locally and leave authentication and permissions unverified. Test, OAuth and automatic refresh are unavailable. A base-token additionally needs an explicit future expiresAt in ISO UTC, within three days. API tokens must omit that field. Never use decoded JWT claims as proof of a grant.

The sixteen actions cover row creation, deletion, retrieval, listing, locking, search, unlocking and updates; base snapshots, metadata and collaborators; adding, listing and removing links; temporary download capabilities and asset upload. Supply existing table/column names and exact 22-character row IDs manually. Columns Ui accepts {"columnValues":[{"columnName":"Enabled","columnValue":false},{"columnName":"Count","columnValue":0},{"columnName":"Title","columnValue":""}]}. These values retain their types when saved. Options accepts only each action's declared keys. Listing/search use an explicit start and a limit of 1–1,000; link lists allow 1–100. Each returns one bounded response, without automatic pages or readbacks.

Every physical stage pauses for a separate signed Human Task assigned to the original run actor, expiring after 15 minutes. Stages can include token issuance, metadata/collaborator reads, the selected operation, file capability issuance, upload and attachment. Review the account, base, method, destination, body and request digest. Current workspace and organization authority, selected account, workflow, input and run must still match. Another Member or Owner cannot approve for the actor. Durable workflow execution, Human Tasks and the SeaTable execution switch must be enabled. API-token exchange explicitly requests a three-day grant; only the selected server's matching response can establish the local grant.

Uploads require an actual owned workspace file with matching content, revision and completed clean scan. Review replacement/append options. An upload can finish before its row attachment is approved; canceling later does not remove the uploaded asset. Temporary download capabilities can disclose file access and are not automatically fetched. Each confirmed stage makes one attempt, including when workflow retries are enabled. An uncertain attempt is not replayed. Inspect the provider before intentionally repeating it; outcomeVerified:false does not verify persistence or delivery. Inputs and selected results can remain in workflow history.

Only ordinary n8n-nodes-base.seaTable imports with explicit typeVersion:2 use this mapping. V1, versionless imports, Tool aliases and triggers need separate migration. Real provider authentication, binary/scan behavior and matching-revision Review walkthroughs remain unverified.

Nextcloud accounts, files and exact requests

Select Basic (nextCloudApi) or an existing OAuth token (nextCloudOAuth2Api) explicitly. Basic uses exactly:

{
  "authentication": "accessToken",
  "accountBaseUrl": "https://cloud.example.org/nextcloud",
  "user": "alice",
  "revision": "1",
  "password": "EXISTING_APP_PASSWORD"
}

OAuth uses the same exact HTTPS installation, username and revision with authentication: "oAuth2", token and a future expiresAt in Unix milliseconds; omit password. Saving and reopening remain local and unverified. Test, login, discovery and refresh are unavailable. Native requests currently require a publicly reachable IPv4 HTTPS installation; localhost and private networks are refused. An optional raw webDavUrl must match this account's supported DAV route exactly.

The 17 ordinary choices cover file copy/delete/download/move/share/upload, folder copy/create/delete/list/move/share, and user create/delete/get/list/update. Supply paths and user IDs manually. Paths begin with /; an empty folder-list path selects the root. Copy and move can overwrite; folder deletion is recursive. Shares can expose data or send mail, and user creation can send invitation or reset mail. Keep only fields for the selected operation: share types use the corresponding recipient; public-link passwords are unavailable on other share types. User lists return one bounded response. Return All does not traverse pages or prove completeness; leave Limit empty when it is enabled.

Each physical request requires a signed Human Task for the original run actor, expiring in 15 minutes, with the exact selected account, destination, method, body and workflow inputs. Current workspace and organization permissions must still pass. Another Member or Owner cannot decide for the actor. Durable execution, Human Tasks and the Nextcloud execution switch must be enabled. One claimed attempt is allowed even when node retries are enabled; inspect uncertain provider effects before deliberately repeating them.

Text uploads preserve an explicit empty body. Binary uploads require an owned Workspace File from a supported upstream binding; downloads are persisted through Workspace Files and its configured scanner. File ownership, readiness, bytes, MIME, hash and full revision must remain current through confirmation and OUTPUT. A direct Nextcloud download can feed an upload; arbitrary other-provider binary outputs are unsupported. A scan-state result does not establish the malware protection of a deployment that uses a test scanner. Large text previews show UTF-8 size and hash; the complete value remains bound to the signed request.

Ordinary n8n-nodes-base.nextCloud mappings cover these operations; this does not establish Tool compatibility or full n8n item, error and output parity. Results remain outcomeVerified: false, with bounded response completeness unknown. Actual provider authentication, installed-version behavior, production file scanning and matching-revision Review remain unverified.

Rundeck API 59 accounts and jobs

Use an existing token with the explicit rundeckApi59 family:

{
  "profile": "rundeck-api59-token",
  "apiBaseUrl": "https://rundeck.example.org/automation",
  "token": "EXISTING_API59_TOKEN"
}

Save the exact public HTTPS installation, including an optional subpath, on port 443. Save and reopen remain local and unverified; Test, login, discovery, version detection and refresh are unavailable. Legacy rundeckApi, API 14, insecure TLS and shell argString settings require explicit migration. The installed server's API 59 support and token rights remain unverified.

Choose job.execute or job.getMetadata and supply the exact Job ID manually. Supported IDs use ASCII letters, digits, underscores and hyphens. Execute Arguments retain the original arguments.arguments collection of unique text name/value rows, including empty, Unicode and "0" values. They become JSON options rather than a shell string. An empty or omitted Filter uses the job/server defaults and can select many remote nodes. A job can run commands, affect infrastructure, notify other systems or incur costs.

Both metadata reads and execution need a separate signed 15-minute Human Task decided by the original run actor. The selected account, exact request, workflow, input, run and current workspace and organization authority must still match. Another Member or Owner cannot approve for that actor. Durable execution, Human Tasks and the Rundeck execution switch must be enabled. Each task permits one physical attempt; retries, polling, output fetch, abort requests and returned-link following are unavailable. Before publishing OUTPUT, the workflow checks the current permissions, approved request and recorded result again. A changed approval or an uncertain result stops OUTPUT. An execution receipt is not job completion. Inspect an uncertain remote effect before deliberately repeating it.

Ordinary version-1 n8n-nodes-base.rundeck imports identify these two operations with an explicit API 59 migration notice. Unknown versions and Tool aliases remain unsupported. Results retain outcomeVerified: false; actual provider acceptance, native sockets and matching-revision Review remain unverified. Request and response bounds are local application policy, not vendor limits.

Rocket.Chat messages and existing accounts

Save an existing account with the rocketchatRestV1Api family:

{
  "profile": "rocketchat-rest-v1-token",
  "domain": "https://chat.example.org/team",
  "userId": "existing-user-id",
  "authKey": "existing-auth-key"
}

Use the exact public HTTPS installation and optional subpath on port 443. Save and reopen are local and unverified. Test, login, discovery, OAuth and refresh are unavailable. Legacy rocketchatApi accounts require an explicit resave with this profile. Saving does not establish the installed server version or message permissions.

The ordinary version-1 n8n-nodes-base.rocketchat node supports chat.postMessage: one channel or room ID, text, Options and attachments. Channel arrays, Tool and Trigger nodes, missing or unknown versions, threads, uploads and arbitrary custom fields are unsupported. Options keep alias, avatar and emoji. Alias and avatar require the server's message impersonation permission. parseUrls defaults visibly to false; attachment and avatar links may still cause Rocket.Chat servers or clients to fetch remote content.

Choose named attachment fields or JSON Parameters. Named fields preserve supported empty strings, false and 0, including nested fields.fieldsValues. JSON Parameters accepts the supported snake_case attachment array. Switching modes in the editor clears the inactive draft; raw or imported inactive fields are refused. At least nonempty text or a supported nonempty attachment is required. The original node declared messageLink but checked messageLinks during execution. Using named messageLink therefore requires explicit messageLinkMigrationAcknowledged: true to map it to message_link; this correction changes the original execution behavior.

Each message needs a signed 15-minute Human Task decided by the original run actor. Review the selected account, resolved fields, destination and exact request before approving. Current workspace and organization permissions, account, workflow, input, run and node must still match. Another Member or Owner cannot approve for that actor. Durable execution, Human Tasks and the Rocket.Chat execution switch must be enabled. Posting may notify users and disclose message data. Each approval allows one physical attempt; automatic retries, redirects, follow-up reads and returned-link navigation are unavailable. Inspect an uncertain remote effect before deliberately repeating it.

The result keeps the bounded complete success: true response, including the message ID and room ID, with outcomeVerified: false. A response is not proof of message delivery. The validated response is saved as a completed physical attempt before the final OUTPUT authority check. Inputs and responses may persist in history. Actual provider acceptance, native sockets, installed-version compatibility and matching-revision Review remain unverified. Request, response and approval bounds are local application policy, not vendor limits.

Elasticsearch REST v8 documents and indices

Select the explicit elasticsearchRestV8Basic account family and enter an existing Basic account:

{
  "profile": "elasticsearch-rest-v8-basic",
  "baseUrl": "https://elastic.example.org/search",
  "username": "EXISTING_USERNAME",
  "password": "EXISTING_PASSWORD"
}

Use the exact public HTTPS installation on port 443, including an optional subpath. This profile currently requires printable ASCII Basic account fields and conservative installation paths; these are application restrictions. Saving and reopening remain local and unverified. Test, discovery, version detection, login, refresh, OAuth, API keys and TLS bypass are unavailable. Legacy elasticsearchApi accounts require explicit migration; saving does not establish the installed version or its read, write or delete permissions.

Five document operations are available: document.get, document.index, document.update, document.delete and document.searchPage. Supply one literal Index ID and, where required, Document ID. Index with a Document ID creates or replaces that document; without one, the server assigns an ID. Server defaults may create an index or run ingest processing. Update merges a partial document and can return noop; Delete removes one document. These operations can disclose data, consume resources and incur costs.

Select parameter schema 2, acknowledge the migration, and keep Simple disabled. Ordinary version-1 n8n imports require this explicit migration to raw response namespaces; source fields cannot overwrite response identity. Missing or unknown versions and Tool aliases remain unsupported. Fields to Send retains imported text values; named controls also accept explicit JSON values such as false and 0. Quote those values when text is intended. Auto Map requires an explicit input object and supports Inputs to Ignore. Inactive fields and unsupported nondefault options are refused by name; remove or repair them before running.

Search Page returns one bounded page with a default limit of 50 and a maximum of 100. Its query accepts only match_all, term or match; source include/exclude filters are supported. Partial search results and timed-out or failed-shard responses are refused. Raw hit metadata and total relation remain available; this is not an all-results export. Bulk, arbitrary query DSL, scripts, upsert, aliases or multiple target syntax, cross-cluster targets, PIT, pagination, refresh and pipelines are outside this implemented subset. A literal name can resolve to an alias on the server; Save does not discover its target, and a returned index identity mismatch is refused.

Each physical read or write needs its own signed 15-minute Human Task approved by the original run actor. The selected account, exact request, workflow, run, input, version, live switch and current workspace and organization authority must still match. Another Member or Owner cannot approve for that actor. One physical attempt is allowed, including when node retries are configured. A complete response is recorded before OUTPUT, which checks current authority and the exact stored result again. An uncertain result is not retried or checked with a hidden follow-up request; inspect remote effects before deliberately repeating it.

All results retain outcomeVerified: false. Successful local simulation does not establish provider acceptance, native DNS/TLS sockets, installed server behavior, real account permissions or matching-revision Review behavior. Request, response, time and structural bounds are application policy, not Elasticsearch vendor limits.

Index operations

Four index operations use the same account and explicit schema-2 migration:

Create, Delete and Get require one literal Index ID. Wildcards, _all, multiple targets, cross-cluster syntax and path expressions are refused. A literal name can still resolve to an alias on the server; saving does not discover its target.

Index Get All defaults to 50 returned namespaces and allows at most 100. The complete response must fit the local 2 MiB budget and pass validation before slicing. The result keeps the selected raw namespaces and coverage.examined, returned, truncated and limit; allResultsComplete and remoteLimitApplied remain false. A short or empty result does not establish a complete inventory. Returned names may include data streams.

Index Create accepts the original empty aliases, mappings and settings defaults and the original default timeout, shard-wait and include_type_name values only as omitted controls. Index Get likewise accepts its original default additional fields only as omitted controls. Server defaults apply and may differ from those original values. Nondefault or unknown fields are refused by name; aliases, mappings, settings and query options are not transmitted. Delete and Get All accept no additional-field or option entries.

False, missing or malformed acknowledgements leave the remote outcome uncertain. Every index read or write still requires the original actor's separately signed request, current workspace and organization authority, one physical attempt and a checked stored result before OUTPUT. An uncertain outcome never triggers an automatic retry, readback or compensating request. Results retain outcomeVerified: false; local simulation does not verify installed-server permissions or remote effects.

ServiceNow incident requests

The ordinary version-1 n8n-nodes-base.serviceNow integration supports five incident operations with explicit Basic authentication and parameter schema 2. Acknowledge migration and keep Simple disabled. Other resources, attachments, OAuth, Tool/Trigger aliases and missing or unknown versions remain unsupported.

Save an existing serviceNowTableV2Basic account using named fields or Advanced JSON:

{
  "profile": "servicenow-table-v2-basic",
  "baseUrl": "https://example.service-now.com",
  "username": "EXISTING_USERNAME",
  "password": "EXISTING_PASSWORD"
}

Use an exact public HTTPS service-now.com tenant origin on port 443, with one subdomain label and no path. Save and reopen are local, make zero provider requests and leave authentication unverified. Test, login, discovery and OAuth are unavailable. Legacy Basic accounts require explicit migration and reselection; saving does not establish tenant version, roles or ACL permissions.

IDs require exactly 32 lowercase hexadecimal characters. Short description is nonempty and limited locally to 160 UTF-8 bytes; Description allows an explicit empty string and at most 4096 characters. Impact and Urgency accept integers 1–3. Unknown and inactive fields are refused by name. Read options allow distinct sys_id, number, short_description, description, impact, urgency and active projections and must include sys_id; display values false, true or all and Exclude reference link are optional. Keep boolean false distinct from an omitted field.

Writes may trigger business rules, notifications and downstream workflows. Lists retain unknown completeness even when short or empty: server limits and ACL filtering do not establish a complete inventory. Raw incident text and responses may persist in run history or reach downstream AI.

Each physical read or write requires its own signed 15-minute Human Task decided by the original run actor. The exact saved account, request, workflow, run, input, version, live switch and current workspace and organization permissions must still match. Another Member or Owner cannot approve for that actor. Durable Human Tasks and the ServiceNow execution switch must be enabled. One physical attempt is allowed, even with node retries; an uncertain outcome is never retried or checked by a hidden readback. Final OUTPUT rechecks authority and the stored result. Results keep outcomeVerified: false. Local simulated transport tests do not verify a real tenant, native sockets or matching-revision Review behavior. Bounds are application policy, not vendor limits.

Gong call reads

The ordinary version-1 n8n-nodes-base.gong integration supports call Get and one-page List through explicit schema-2 migration. Transcript is a separate native read, with its own approval; selecting transcript inside an imported Get does not cause a hidden request. Acknowledge the raw-response migration and select the matching saved account. Tool aliases, unknown versions, automatic cursors, media retrieval, users and OAuth are outside this implemented subset.

Save an existing gongRestV2Basic account with Access Key and Access Key Secret:

{
  "profile": "gong-rest-v2-basic",
  "baseUrl": "https://api.gong.io",
  "accessKey": "EXISTING_ACCESS_KEY",
  "accessKeySecret": "EXISTING_ACCESS_KEY_SECRET"
}

This profile supports only the explicitly selected https://api.gong.io origin on port 443. It does not discover the customer's region, tenant, plan, scopes, key expiry or trusted-IP policy. Legacy gongApi and OAuth accounts need explicit migration and reselection. Save and reopen are local and unverified, with zero provider requests; Test, discovery, login and refresh are unavailable.

Use exact decimal ID strings of 1–20 digits, preserving leading zeros; numeric IDs and automatic locator searches are refused. List filters allow at most five distinct Call IDs or Primary user IDs. Times use UTC seconds, for example 2026-10-21T10:00:00Z, with From before To. Get supports Include fields for parties, topics, brief, public comments, highlights, key points, call outcome, outline, structure and trackers; List supports parties and topics. Named and JSON controls preserve explicit false. Unsupported fields are refused rather than silently dropped.

All three POST requests are reads. The complete bounded raw page can contain participants, public comments, analysis or transcripts, including records beyond the local selected-item limit. Those raw namespaces may persist or reach downstream AI; a local limit is not a server-side privacy limit or a complete export. The local response budget is 1 MiB, with a 30-second transport bound; these are application restrictions.

Each read requires the original actor's separately signed 15-minute Human Task, the exact selected account and request, and current workspace and organization authority. Durable Human Tasks and the Gong execution switch must be enabled. One physical attempt is allowed, including with node retries. Settlement and final OUTPUT authenticate the stored response and receipt; an uncertain read is not retried, refreshed or followed by a hidden request. Results keep outcomeVerified: false. Provider acceptance, installed permissions, native sockets and matching-revision Review remain unverified.

Okta user requests

The ordinary version-1 n8n-nodes-base.okta integration supports five user operations through explicit parameter schema 2 and manual SSWS authentication. Acknowledge migration and select the matching saved account. OAuth, unknown versions, Tool/Trigger aliases, password changes, lifecycle operations and custom profile attributes are outside this implemented subset.

Save an existing oktaSswsUsersV1 account using named fields or Advanced JSON:

{ "profile": "okta-ssws-manual-v1", "origin": "https://example.okta.com", "accessToken": "EXISTING_SSWS_TOKEN" }

Use the exact public HTTPS tenant origin on port 443, without a trailing slash, path, query or fragment. Save and reopen are local and leave authentication unverified, with zero provider requests. Test, login, OAuth, discovery and refresh are unavailable. Legacy accounts require explicit migration and reselection; saving does not establish token permissions or tenant policies.

Manual user IDs start with 00u and use the supported ASCII form; me, login lookup and locator searches are refused. Profile controls support the 31 named standard fields, including Postal address and Zip code. Required First name and Last name retain the API's explicit null option. Null, empty text and omitted fields remain distinct where supported. Duplicate create fields, unknown attributes and expressions are refused. A Link header is untrusted data: the application never follows it automatically. A short page is not a complete inventory.

Raw user profiles may persist in run history or reach downstream AI. Each physical request requires the original actor's separately signed 15-minute Human Task, the exact selected account and request, and current workspace and organization permissions. Another Member or Owner cannot approve for that actor. Durable Human Tasks and the Okta execution switch must be enabled. There is one physical attempt, including with node retries; uncertain outcomes do not cause a hidden readback or a second delete. Final OUTPUT authenticates the stored response and receipt and rechecks authority. Results keep outcomeVerified: false. The local response budget is 1 MiB with a 30-second transport bound. Simulated tests do not verify a real tenant, provider permissions, native sockets or matching-revision Review behavior.

Unleashed sales orders and stock

The ordinary version-1 n8n-nodes-base.unleashedSoftware integration supports one sales-order page, one product's stock and one stock page through explicit parameter schema 2. Acknowledge migration and select the matching company account. Automatic pagination, writes, unknown versions, Tool/Trigger aliases and unsupported filter values are outside this implemented subset.

Save an existing unleashedHmacCompanyV1 account using named fields or Advanced JSON:

{
  "profile": "unleashed-hmac-sha256-query-manual-v1",
  "apiId": "EXISTING_API_ID",
  "apiKey": "EXISTING_API_KEY",
  "clientType": "confluye/local"
}

The origin is fixed to https://api.unleashedsoftware.com on port 443. Client type uses two lowercase letter, digit or hyphen segments separated by /, with at most 64 characters. The server signs the exact outgoing query with the company's API key. Save and reopen are local and unverified, with zero provider requests; Test, discovery, login, OAuth and refresh are unavailable. Saving does not verify company access, subscription or permissions.

Pages default to 1 and allow 1–1,000,000; limits default to 100 and allow 1–100. These are local restrictions. Keep Return all disabled. Sales filters include Customer ID GUID lists, Customer code, End date, Modified since, Order number, Order status and Start date. An Order number cannot be combined with other filters because the provider overrides them. Status supports Backordered, Completed, Deleted, Parked and Placed; omitting it does not include deleted orders automatically.

Stock filters include As-at date, Is assembled, Modified since, Product ID GUID lists, Warehouse code and Warehouse name. Explicit false is preserved. Order by supports only the empty default until a closed column contract is available. GUID lists preserve order and duplicates and allow up to 50 entries. Dates require real YYYY-MM-DD values; imported date-time truncation is refused. Raw .NET date representations remain raw, and a missing stock record does not establish zero stock.

Customer, order, stock and cost data may persist or reach downstream AI. Each read requires the original actor's separately signed 15-minute Human Task, the exact company account and request, and current workspace and organization authority. Durable Human Tasks and the Unleashed execution switch must be enabled. One physical attempt is allowed, including with node retries; there is no automatic next-page, count, retry or readback request. Settlement and final OUTPUT authenticate the stored response and receipt. Results keep outcomeVerified: false. The local response budget is 1 MiB with a 30-second transport bound. Provider acceptance, native sockets, company permissions and matching-revision Review remain unverified.

AWS SQS and Lambda requests

AWS SQS supports one sqs.sendMessage request; AWS Lambda supports one lambda.invoke request. Ordinary version-1 imports require explicit parameter schema 2 and migration acknowledgement. These selected actions do not provide complete AWS credential, region, binary or node parity.

Save an existing awsSqsManualSigV4V1 or awsLambdaManualSigV4V1 account using named fields or Advanced JSON. Use the matching profile, aws-sqs-manual-sigv4-v1 or aws-lambda-manual-sigv4-v1, partition aws, a supported region, the declared 12-digit AWS account ID, Access key ID and Secret access key. Choose manual-static, or manual-temporary with an existing Session token and explicit UTC expiresAt. Temporary credentials must remain valid beyond the 30-second request budget plus a 60-second safety margin; replace and approve them again when necessary. The declared account is an input restriction, not a verified AWS identity.

Save and reopen send zero provider requests and leave authentication unverified. There is no connection Test, sign-in, STS, role assumption, system credential chain or refresh. The fixed endpoint map includes 34 commercial regions per service. GovCloud, China, isolated partitions, custom, FIPS and dual-stack endpoints are unsupported; the map does not promise every current AWS region.

SQS supports the selected string and canonical decimal number attributes, preserves omitted delay versus zero and validates returned body/attribute MD5 values. FIFO requires explicit group and deduplication IDs and refuses per-message delay, including zero. Binary attributes are refused. Lambda does not accept an empty payload, binary results, logs or client context. Function versions, aliases and $LATEST can change without a configuration read. Messages and invocations may incur charges.

Every request needs the original actor's separately signed 15-minute approval, exact saved account and request, and current workspace and organization authority. Durable Human Tasks and the service execution switch must be enabled. SigV4 date and signature are generated at physical dispatch; approval binds the stable request and saved account revision. There is one physical attempt, with no automatic resend or readback after an uncertain result. Settlement and final OUTPUT check the authenticated stored response receipt. outcomeVerified: false does not establish downstream completion.

Business inputs and bounded JSON responses can persist in workflow history and reach downstream AI. Local limits are 32 KiB per request, 1 MiB per response and 30 seconds; these are application restrictions, not AWS limits. Local simulated tests do not establish actual AWS permissions, endpoint acceptance, native sockets or matching-revision Review behavior.

ServiceNow fixed-table ID lookups

Alongside the five incident actions, ServiceNow supports six first-page ID lookups using the same serviceNowTableV2Basic account and explicit schema-2 migration:

Explicitly enable lookupIdProjectionAcknowledged. This migrates the original Get all action to a restricted first page of sys_id values only. The limit is 1–50, default 50, with offset zero, display values disabled and reference links excluded. There is no custom query, arbitrary table, count request or automatic pagination. Empty and short pages do not establish a complete inventory. IDs can still disclose organizational structure and are retained under ordinary history and downstream-data rules.

Only HTTP 200 JSON with the exact result envelope and rows containing one lowercase 32-hex sys_id is accepted. Extra fields and display data are refused before public output. Each lookup requires the original actor's signed approval, current workspace and organization authority and one physical claim. Aborted or uncertain lookup requests settle without automatic replay; inspect the provider separately before deciding to repeat. Attachments, arbitrary table-record actions, user actions and OAuth are outside this subset. These six lookups have local simulated coverage; real provider permissions, native sockets and matching-revision Review remain unverified.

Twist text and JSON actions

The ordinary version-1 Twist integration supports 22 selected text/JSON actions through explicit parameter schema 2 and migration acknowledgement:

Save a twistManualBearerV1 account using named fields or Advanced JSON:

{ "profile": "twist-manual-bearer-v1", "accessToken": "EXISTING_MANUAL_TOKEN" }

This is a partial manual-token authentication profile. The captured official documentation provides testing tokens and requires OAuth for production public integrations. Original OAuth credentials do not migrate automatically. Saving and reopening are local and unverified, with zero provider requests; Test, sign-in, discovery, OAuth acquisition and refresh are unavailable. The endpoint is fixed to https://api.twist.com/api/v3 on port 443. Saving does not establish token identity, workspace membership or permissions.

Use exact parent/workspace/recipient IDs and literal text or structured values. Expressions are unsupported. Supported controls preserve omitted fields, empty text, false and zero where permitted. Timestamps require explicit Unix-second integers; timezone-dependent date conversion is not reproduced. Channel Update requires a nonempty Name. Temporary IDs must be negative; an imported default of zero requires repair. Original uppercase ordering requires lowercase asc or desc, and the deprecated is_starred filter requires an explicit current boolean. Conversation-list additional fields must be moved to native Filters. Conversation Create retains the original workspace ID although the current endpoint documentation does not list it; provider acceptance remains unverified.

Lists require singlePageRawAcknowledged: true; keep Return all disabled. Limits default to 50 and allow 1–500. Channel List receives and examines the complete bounded server array before local slicing; the local limit does not prevent other received records from entering raw history. Other lists make one bounded request. No list proves complete history or follows another page. Comment Get requires explicitly choosing the documented object or original comment-wrapper response shape. Raw HTTP bytes and the native data projection remain separately available; bodyless 200/204 responses and outcomeVerified: false do not prove an external effect.

Creating, editing, deleting, archiving or unarchiving content can change access or notify people. Exact recipients and parent IDs matter; raw user/message data can persist in history or reach downstream AI. Every read and write requires the original actor's separately signed 15-minute approval, the exact saved account and request, and current workspace and organization authority. Durable Human Tasks and the Twist execution switch must be enabled. One physical attempt is allowed, with no automatic retry, reconnect, cursor following or hidden readback after uncertainty. Final OUTPUT validates the signed stored response receipt and current authority.

Attachments and binary input for comment, conversation-message and thread Create/Update are unavailable. Nonempty binary references and attachment arrays are refused; no URL is fetched as a substitute. Action-button links remain provider payloads and are not fetched by this integration. Local limits are 64 KiB request, 1 MiB response and 30 seconds, not provider limits. Local simulated checks do not verify production OAuth, provider DNS/TLS/HTTP, actual encrypted database Save/reopen, the full persisted original-actor approval flow, human MFA or matching-revision Review.

SyncroMSP REST: explicit migration and one approved request

SyncroMSP supports 20 ordinary actions: create, delete, get, update and one-page list for contacts, customers and tickets, plus create, delete, get, mute and one-page list for RMM alerts. Select an existing syncromspBearerRestV1 account with profile syncromsp-bearer-manual-v1, a lower-case ASCII tenant label and an existing bearer token. The destination is fixed to https://tenant.syncromsp.com/api/v1 for that label. Saving and reopening the account is local and unverified; it sends no provider request. Test, sign-in, OAuth, token refresh and automatic discovery are unavailable. Authentication, tenant identity, permissions and provider acceptance still need verification.

Original numeric n8n node version 1 imports preserve the original parameters for repair. They require manual schema-2 conversion, a separately selected supported account, wireMigrationAcknowledged:true and the account's queryAuthMigrationAcknowledged:true: the original query API key is not silently reused as a bearer token. Tool and Trigger identities and other node versions are unavailable. Expressions and unsupported address-array writes remain visible for explicit repair and cannot execute as supported fields.

Review the exact native field names. Customer name and address aliases use their explicit wire names; ticket fields retain explicit empty, null, false and zero values where the schema permits them. An omitted invoice-CC field stays omitted, while an explicit empty list clears it. Updates require at least one selected field. For example, an explicitly migrated customer update can preserve these distinct values:

{
  "parameterSchemaVersion": 2,
  "wireMigrationAcknowledged": true,
  "id": 17,
  "fields": {
    "notes": "",
    "notification_email": null,
    "get_sms": false,
    "invoice_cc_emails": []
  }
}

RMM mute additionally requires muteQueryMigrationAcknowledged:true; it sends the duration in the documented query location rather than the original request body. The retained flat RMM-create body and ticket asset_id/ticket_type fields differ from current documentation; their live compatibility remains unverified. No alternate request is sent automatically to resolve those differences.

Lists require singlePageAcknowledged:true, default to a local limit of 25 and allow at most 100 returned items. The bounded raw page is retained before local slicing; examined and returned counts do not establish completeness. returnAll:true, cursor/Link following and hidden readback are unavailable. Local limits are 32 KiB per request, 1 MiB per response and 30 seconds per invocation. These are application limits, not provider guarantees; IPv6-only destinations are unsupported.

Every read and write needs the original run actor's separate exact-request confirmation and current workspace and organization authority. Approval expires after 15 minutes and is tied to the selected account, workflow, inputs and request. Changing the saved account requires new consent. Contact, customer, ticket and alert data can remain in execution history or reach downstream AI. Review the destination, record IDs, literal body and applicable changes before confirming. One physical attempt is allowed; ambiguous results are not retried automatically. Bodyless HTTP 200/204 and returned IDs do not verify the remote effect; results retain outcomeVerified:false.

Local controls, encryption, import round trips and the production gateway have been exercised with simulated storage and provider transport. Matching Review walkthroughs, current composed database durability, native provider sockets, real authentication/effects and AI-provider interpretation are not established by those local checks.

Taiga: manual token and explicit field migration

Taiga supports create, delete, get, update and one-page list for epics, issues, tasks and user stories. Select taigaBearerManualV1 with profile taiga-bearer-manual-v1, an exact public HTTPS origin and an existing accessToken. Named fields and Advanced JSON use the same encrypted account format. Saving and reopening are local and unverified. The original username/password login requires explicit migration and account reselection; no login, refresh, credential Test or option discovery occurs.

Original ordinary imports retain their type, version, parameters and credential references for repair. Native requests require parameterSchemaVersion:2 and migrationAcknowledged:true. Tool and Trigger identities are separate. Enter IDs manually. Updates require the current record version and manualVersionAcknowledged:true; version 0 is valid and no version lookup runs. Selected assigned_to or milestone null values clear the field where supported, while omission leaves it unchanged. False, empty descriptions or blocked notes and empty tag lists retain their distinct values. Corrected filter names, comma-separated query tags and hash-prefixed epic colors need explicit migration. Legacy user-story type additionally needs unverifiedLegacyTypeAcknowledged:true; live support remains unverified.

Lists need singlePageRawAcknowledged:true, default to a local limit of 50 and allow at most 100 selected items. The bounded raw page remains available; completeness is unknown. Return all, automatic continuation and following server next URLs are unavailable. Local limits are 64 KiB per request, 1 MiB per response and 30 seconds per invocation; these are application limits rather than provider guarantees.

Twake: historical n8n plugin message

Twake supports the ordinary message.send action through twakePluginWorkspaceKeyV1, using profile twake-plugin-workspace-key-v1 and an existing workspaceKey. The fixed destination is https://plugins.twake.app/plugins/n8n/actions/message/save. Basic, JWT and Twake Server accounts are separate unsupported modes. Saving is local and unverified; the historical plugin's current availability, permissions, response status and wrapper have not been verified with a live account.

Enter the channel ID and formatted content, with explicit schema-2 migration. The request sets object.channel_id, object.content.formatted and object.hidden_data.allow_delete:"everyone". Nonempty sender name or icon overrides are forwarded; empty overrides are omitted. An icon URL is forwarded as data and never fetched locally. One native invocation produces one request and one result, explicitly replacing the original multiple-input last-result behavior.

Both providers require the original run actor's separate exact-request confirmation, valid for 15 minutes, plus current workspace and organization authority. Changing the saved account, ciphertext, workflow, inputs or request requires fresh consent. Raw descriptions, messages and personal data can remain in history or reach downstream AI; remote changes can notify recipients or incur costs. One physical attempt is allowed. Ambiguous results are not automatically resent or read back. A response, returned ID or bodyless deletion does not verify the remote effect; outcomeVerified remains false.

The result envelope retains the bounded raw UTF-8 body, digest and item counts. Contract checks require Taiga create 201, delete 204 with no body, other Taiga responses 200 and Twake send 200; these expectations are not live-provider acceptance evidence. Invalid encoding, leading BOM, duplicate JSON keys, oversized bodies and reflected account secrets are refused. Local storage, request, approval and runtime tests use simulated database facts and provider transport. Current composed PostgreSQL durability for these providers, native sockets, real authentication/effects, AI interpretation and matching Review walkthroughs remain untested.

Telegram: ordinary Bot API actions

The native Telegram profile provides 23 synchronous ordinary actions: six chat operations, two callback operations, file metadata, and fourteen message operations. Choose the existing bot token account telegramBotTokenPublicV1, enable the workspace live transport, then select that account on the block. Named fields and Advanced: edit JSON save locally, encrypted and unverified; saving sends no HTTP or DNS request. Test, token acquisition, OAuth and token refresh are unavailable for this profile.

{ "profile": "telegram-bot-token-public-v1", "accessToken": "EXISTING_BOT_TOKEN" }

Requests use the public Bot API at https://api.telegram.org. Custom/local Bot API servers require a separate profile and account migration. Existing alert credentials and queued chat notifications have a separate lifecycle; their account selection and delivery references do not verify this ordinary-action profile. Incoming triggers retain their separate publishing and webhook behavior.

Repair an imported ordinary node explicitly: retain its original version, parameters and credential references, select the supported account, set parameterSchemaVersion: 2, acknowledge wireMigrationAcknowledged: true, and supply the original numeric sourceVersion (1, 1.1 or 1.2). Import preserves unsupported, hidden and nested fields for repair; execution refuses them rather than silently dropping them. Ordinary support does not extend to Tool, HitlTool or Trigger aliases.

File metadata requires download: false. Six single-media actions require binaryData: false and either an existing bot file_id or a URL Telegram fetches. The application does not download those URL bytes, and file_unique_id cannot replace file_id. Media groups retain two to ten photo/video references. Binary upload/download and sendAndWait are unavailable in these actions. Existing application response waits use a different lifecycle and do not establish support for the imported external-responder flow.

Preserve typed zero coordinates and false notification values; an empty chat description clears it. Editing uses either the chat/message identity or inline-message identity according to the selected mode. Reply markup retains the source rows/row/buttons structure; inline-query results use a bounded JSON string and the supported closed variants. Message attribution and link-preview defaults follow the original version. Removing the original branding requires the explicit remove-n8n-attribution-v1 migration; retaining it requires the original instance ID.

Review the exact chat, user, callback, text, location, keyboard, media URL and file-ID disclosure before approving. Only the original run actor can approve that exact request within 15 minutes. Current workspace and organization access, saved account revision and request inputs are checked again before sending and before OUTPUT. Each invocation makes one physical attempt. A response does not prove later delivery, and an uncertain outcome is never automatically resent, refreshed, redirected or read back.

Output retains bounded raw response text and original source items. HTTP 200 with ok: false fails. Administrators are unwrapped into source items; a media-group array remains inside its original envelope. Identities, messages and location data can remain in history and reach downstream AI. Local production-path tests use simulated storage and provider responses; matching-revision Review, real DB/Redis durability, native sockets, provider acceptance and human approval remain unverified.

KoBoToolbox: forms, submissions, hooks and file metadata

The native KoBoToolbox profile provides 17 ordinary JSON/metadata actions: form get/list page/redeploy; submission get/list page/delete/get validation/set validation; hook list page/get/get logs/retry all/retry one; and file list page/get/delete/create by URL reference. Choose the instance owning the existing token: Global, EU or one exact public self-hosted HTTPS origin on port 443. IPv6-only transport is unsupported.

{
  "profile": "kobo-token-public-instance-v1",
  "instance": "global",
  "origin": "https://kf.kobotoolbox.org",
  "originNormalizationAcknowledged": true,
  "token": "EXISTING_ACCOUNT_TOKEN"
}

The saved credential type is koboTokenPublicInstanceV1. EU uses instance: "eu" and https://eu.kobotoolbox.org; self-hosting uses instance: "selfhost" and the exact approved public origin. Named fields and Advanced: edit JSON save the existing token locally, encrypted and unverified, with no HTTP or DNS request. Test, form discovery, dynamic options, token acquisition, OAuth and refresh are unavailable. Enter exact form and resource IDs manually. Normalizing a legacy URL requires explicit acknowledgment and account reselection.

Imported ordinary nodes retain their original parameters, version and credential references for explicit repair. Native inputs require parameterSchemaVersion: 2, migrationAcknowledged: true and rawRetentionAcknowledged: true. Unsupported or inactive fields remain in the source ledger and refuse execution until repaired. Source getAll aliases migrate to List page with singlePageAcknowledged: true and an explicit limit from 1 to 1,000. Source form/hook fallback is 1,000; submission UI default 100 differs from its runtime fallback of 1,000. File/log lists use a local cap. One page and its next/count data never establish completeness; the application does not follow Next automatically.

file.get requires download: false and returns metadata. file.create accepts fileMode: "url" plus fileUrl; it sends an inert URL reference to KoBo without fetching it locally. Uploading content, downloading attachments and registering triggers require separate capabilities and are unavailable. Tool/Trigger aliases do not inherit ordinary support.

Submission transformation preserves source slash-group trimming, wildcard masks, select-after-number precedence and longitude-first GeoJSON. Coordinate zero retains the source truthiness behavior. Prototype paths, bracket-path syntax and transformations producing non-JSON numbers require repair. Hook log mode source-start-end preserves source query names; current-dates explicitly maps ISO dates to start_date/end_date. Hook retries can resend submissions to provider-configured destinations and cause notifications, retries or costs; the application does not discover those destinations or verify delivery.

The original run actor must approve the exact request within 15 minutes, with current workspace and organization access and the selected token/origin revision. Changing the account, workflow, inputs or access invalidates pending approval. Each invocation makes one physical attempt; an unknown outcome is never automatically resent. A successful HTTP response does not verify deletion, redeployment or downstream delivery.

Outputs preserve bounded raw UTF-8 response text, named sourceItemsJson, original wrappers and examined/returned counts. The file-list nested array stays nested. A submission-deletion success: true value is source synthesis after an accepted response, while outcomeVerified stays false. Raw submissions and hook logs can contain identities, locations, payloads and downstream secrets; retained data can appear in history or reach AI. Local tests use simulated storage, DNS/socket facts and provider responses. Matching-revision Review, real DB/Redis durability, native sockets, provider acceptance and human approval remain unverified.

AWS Comprehend, SNS and SES requests

AWS Comprehend supports three text analyses; SNS supports topic create, delete and publish; SES supports six custom verification email actions, two email sends and five template actions. These nineteen ordinary version-1 imports require parameter schema 2, explicit migration acknowledgement and an acknowledgement that inputs and operation data persist in workflow history and may reach downstream AI or remote actions. The SNS Trigger lifecycle, Tool variants, extra AWS APIs, attachments and raw MIME are unsupported.

Select the exact manual credential family: awsComprehendManualSigV4V1, awsSnsManualSigV4V1 or awsSesManualSigV4V1. Their profiles are respectively aws-comprehend-manual-sigv4-v1, aws-sns-manual-sigv4-v1 and aws-ses-manual-sigv4-v1. Named fields and Advanced JSON save the same strict account. Supply partition aws, an explicitly supported commercial region, a declared 12-digit account ID, mode, access key ID and secret access key. manual-temporary additionally requires a supplied session token and UTC expiry beyond the 30-second invocation budget plus a 60-second margin. manual-static forbids token and expiry. Save/reopen is local and unverified, performs zero provider HTTP and does not reveal saved secrets. Test is unsupported and bulk Test skips these three families. No STS, role assumption, system provider chain, refresh, account probe, discovery or picker read occurs. The account ID restricts inputs and does not prove AWS identity.

Each run requires a separate exact-request HumanTask approved by the original run actor within fifteen minutes. Another administrator cannot substitute. Current workspace and organization membership, policy, workflow/node/input/account revisions, preference generation, consumed task/wait and lease are checked again at execution and OUTPUT. One checkpoint claim allows at most one physical request. An error, timeout, disconnect, lost settlement or missing receipt leaves an uncertain outcome and never grants automatic resend. HTTP acceptance, MessageId or TopicArn does not prove delivery, processing, verification, topic deletion, remote template state, privileges or identity. Comprehend discloses text; SNS can fan out to existing subscribers; SES may send mail or change remote templates. Charges and downstream effects are possible.

Exact bounded raw response bytes are application-encrypted into WorkspaceFiles and bound to the run, actor, account, request, task/wait and HTTP status. OUTPUT reopens and authenticates that protected artifact under current authority. This does not claim raw database encryption, inspection of plaintext by antivirus, or completed backup/export deletion verification. The configured file scanner sees ciphertext; unavailable storage/scanning prevents a usable result without resending. Operation data, including addresses, templates and other personal information, remains in ordinary workflow history and can reach downstream AI and remote actions. Require rawRetentionAcknowledged:true before execution; do not supply data you cannot retain or disclose through those paths.

These profiles impose local limits below provider limits: 4096 UTF-8 bytes per text/template field, 16 KiB request, 1 MiB response, thirty seconds, sixty-four template-data rows and fifty total destination recipients. Lists are exactly one bounded page, default twenty, at most fifty custom verification templates or one hundred ordinary templates. Require boundedPageAcknowledged:true, returnAll:false and explicit opaque continuationToken for a subsequent separately approved page. No hidden pagination, readback or retry occurs. HTML, redirect and template strings remain inert locally and are never fetched. The supported endpoint maps contain 12 Comprehend, 34 SNS and 27 SES commercial regions. Available regions depend on the selected service.

SNS names and ARNs are bound to the selected commercial region and declared account. FIFO create/delete accepts .fifo; FIFO publish is refused because the ordinary source lacks MessageGroupId. Comprehend retains ordered language/entity arrays, zero scores and numeric spelling in protected raw data; dominant-language simple mode also provides a derived map. SES custom verification update repairs failureRedirectionURL and fromEmailAddress; template HtmlPart is controlled by its own presence, including HTML-only updates. Unknown/hidden original values and original credential references remain in the import source ledger until deliberately repaired. Corrected form encoding replaces the legacy raw query concatenation and JSON-quoted form body. Import, repair, serialize, reopen and reimport do not grant account or provider verification.

The exact ordinary operation IDs are:

• sns.create
• sns.delete
• sns.publish
• comprehend.text.detectDominantLanguage
• comprehend.text.detectSentiment
• comprehend.text.detectEntities
• ses.customVerificationEmail.create
• ses.customVerificationEmail.delete
• ses.customVerificationEmail.get
• ses.customVerificationEmail.getAll
• ses.customVerificationEmail.send
• ses.customVerificationEmail.update
• ses.email.send
• ses.email.sendTemplate
• ses.template.create
• ses.template.delete
• ses.template.get
• ses.template.getAll
• ses.template.update

For example, an HTML-only SES template update uses:

{
  "parameterSchemaVersion": 2,
  "migrationAcknowledged": true,
  "rawRetentionAcknowledged": true,
  "templateName": "existing-template",
  "updateFields": { "htmlPart": "<p>Reviewed replacement content</p>" }
}

HaloPSA: existing bearer and protected results

HaloPSA supports create, delete, get, update and one-page listing for clients, sites, tickets and users. Select an existing haloExistingBearerV1 account with profile halo-existing-bearer-v1, exact public HTTPS origin, explicit API basePath, existing bearer accessToken and operator-supplied expiresAt. Named fields and Advanced JSON save the same encrypted, write-only account. Save and reopen do not contact HaloPSA or verify the token. Connection Test, token acquisition, OAuth, refresh and option discovery are unavailable. Private or on-premise instances, alternate ports and IPv6-only instances need separate networking support.

Import ordinary numeric-version-1 n8n-nodes-base.haloPSA nodes, reselect the account, and review the explicit schema-2 migration. Each action requires migrationAcknowledged, rawRetentionAcknowledged and onePhysicalAttempt; these acknowledgements start unchecked. Enter resource IDs manually. A list also requires singlePageAcknowledged, page_no and page_size (1–1000). A supplied source limit must match page_size. getAll returns one page and does not establish completeness. The request uses HaloPSA's pageinate spelling. Client, site and user active-status filters map to the current include flags; the imported ticket active-status filter must be removed during repair because the current Tickets contract has no corresponding flags.

To set a user password, save optional userPassword in the selected encrypted account and use password:"saved-account-password" in the selected create/update fields. The request maps that reference to Users.new_password. Plaintext passwords in ordinary n8n input or serialized workflow parameters are rejected before import persistence. Replacing the saved password invalidates existing approvals.

Review the authored fields before approving their protected request digest. Only the original run actor can approve, within fifteen minutes; current workspace and organization execution access, selected-account revision, workflow/input identity and consumed task/wait remain required. One checkpoint claim permits one physical request. A timeout, disconnect or missing result does not authorize an automatic resend. Tenant automation may send ticket mail, incur charges or perform irreversible deletion. HTTP acceptance does not confirm those effects.

Request bodies, personal data, raw responses and bounded provider errors use purpose- and context-bound encrypted checkpoint envelopes. Public outputs expose protected cardinality and digests. Execution history projections and portable exports omit protected request content; re-enter required content and reselect the account after importing an export. Encrypted payload remains while its checkpoint exists: no automatic checkpoint deletion or retention deadline is promised. This profile does not provide a WorkspaceFiles download or antivirus guarantee. Local limits are 64 KiB input, 1 MiB request/response and thirty seconds per request. See the operations and examples.

GoToWebinar: existing bearer and protected responses

GoToWebinar exposes 23 ordinary actions for attendees, coorganizers, panelists, registrants, sessions and webinars. Use an existing entitled app/account and a manually supplied bearer with the required current collaboration permissions. Account entitlement, token validity and the current authorization-server transition remain external prerequisites. This integration does not acquire or refresh tokens.

Save an account of type gotoWebinarExistingBearerV1 with the same profile, existing accessToken, actual UTC expiresAt, and decimal-string accountKey and organizerKey. Named fields and Advanced JSON save the same encrypted, write-only account. Save/reopen is local and unverified, with no provider HTTP/DNS. Connection Test, OAuth, refresh, account lookup and dynamic option discovery are unavailable; bulk credential tests skip the account. Resource requests use fixed https://api.getgo.com:443/G2W/rest/v2.

Every action uses explicit schema 2 with migrationAcknowledged and rawRetentionAcknowledged; acknowledgements start unchecked. Enter IDs manually as decimal strings, including values above 2^53. Lists require singlePageAcknowledged and read one explicit page; a next page needs a separate request and approval. Original returnAll does not start a loop. Plain attendee/coorganizer/panelist arrays retain their order without local slicing, and completeness stays unknown. Registrant lists require page/limit; session and webinar lists require page/size and their selected webinar or explicit date range.

Review recipients, registration content, survey answers, invitation/cancellation emails, notifications, permanent or whole-series deletion, quota and costs. Only the original persisted run actor can approve the exact request within fifteen minutes; current workspace AND organization execution rights, selected account, published node/input, task/wait identity and execution lease remain required. One durable claim permits at most one physical request. Timeout, disconnect, uncertain settlement or unavailable output storage never triggers resend, token refresh, redirect or readback. HTTP acceptance does not establish delivery, deletion completion or billing state.

Raw provider responses, including participant and join-link data, are encrypted in protected Workspace Files. Public results contain authenticated file references, digests and counts. Scanning covers ciphertext only; there is no plaintext antivirus assurance or new automatic retention deadline. This output cannot directly map participant or registration business fields to later nodes, and this integration adds no decrypt/download viewer. Authored inputs and non-secret import ledgers can still contain personal data in ordinary workflow configuration and approval/checkpoint history; this is not raw database encryption.

Webinar PUT updates require the complete reviewed session definition and completeSessionDefinitionAcknowledged; omitted fields may be replaced, with no hidden GET to fill them. single_session and series are supported; source sequence recurrence is refused until its conflicting request schema is resolved. Local bounds are 64 KiB input, 1 MiB request/response and thirty seconds, rather than vendor limits. See all operation IDs and examples.

Clearbit and Medium: conditional existing access

Clearbit provides company autocomplete, company enrichment and person enrichment. Autocomplete uses a saved clearbitPublicAutocompleteV1 policy account with publicAsIsPolicyAcknowledged:true; it has no key field and sends no Authorization header. Enrichment requires clearbitLegacyBearerV1, an existing accessToken, existingEligibleAccountKeyPlanAcknowledged:true and currentStreamingSchemaUnverifiedAcknowledged:true. Existing pre-2024 account/key/plan eligibility must be established separately; exact current streaming schemas remain unverified.

Medium provides post.create for an explicit author or publication and publication.getAll for an explicit author. Save mediumExistingIntegrationTokenV1 with an existing accessToken, explicit tokenKind (integration or oauth), existingPermittedIntegrationTokenAcknowledged:true and unsupportedSchemaAndVendorAcceptanceAcknowledged:true. An OAuth-issued bearer also needs its actual future expiresAt; there is no refresh. Current policy does not permit new integrations/tokens, and a new host is not automatically an existing permitted integration. Archived API schemas and vendor acceptance remain unverified.

Named fields and Advanced JSON save the same encrypted, write-only account locally. Save/reopen remains unverified and sends no provider HTTP/DNS; Test, OAuth setup, token acquisition/refresh, discovery and hidden identity lookup are unavailable. Bulk credential tests skip these profiles. Account acknowledgments start unchecked and do not verify entitlement or permissions.

Every action requires schema 2 and explicit migrationAcknowledged, onePhysicalAttempt and rawResponseRetentionAcknowledged. Review the actual input and request as the original run actor with current workspace AND organization execution authority. The selected account revision, published node/input, original task ID AND wait token and execution lease remain bound through one winning physical claim and fresh authenticated output. Uncertainty after sending never triggers automatic resend, refresh, readback or redirect.

Results expose actual bounded business JSON for downstream mapping, plus raw UTF-8 response and source-item ledgers. Those fields can contain personal or post data in ordinary workflow/run/checkpoint/HumanTask history and may reach downstream nodes or AI. Credentials use existing encryption; this integration adds no raw-history encryption, automatic checkpoint expiry, protected-file transfer or scanner guarantee. Clearbit HTTP 202 is pending, without polling. Medium HTTP acceptance does not prove publication, delivery or follower notification. See exact actions and migration controls.

FileMaker: existing Data API session

FileMaker offers eight ordinary node-version-1 actions: get one record, get one page, find, create, edit, perform a script, duplicate and delete. This is an explicit migration to an existing manual Data API v1 session. Original username/password login/action/logout behavior, v2 and vLatest are unsupported.

Save a filemakerManualSessionV1 account with the exact public HTTPS origin, database, existing session token and operator asserted UTC expiry. A session may expire after 15 minutes idle; the supplied expiry does not verify it. Named fields and Advanced JSON save the same encrypted, write-only account locally. Save/update/reopen sends no provider request and leaves the account unverified. Test, login, logout, refresh, identity lookup and layout/field/script discovery are unavailable; bulk tests skip this profile.

Enter a manual layout schema and identifiers, then separately review the resolved request as the original run actor with current workspace AND organization execution authority. Selected scripts can cause external effects even on GET/find. One physical request is allowed; uncertain outcomes never trigger automatic retry, readback or session renewal. Useful business fields, portals, raw response and source items can remain in ordinary workflow/checkpoint history and reach downstream nodes or AI. This adds no automatic expiry, protected-data viewer, raw-database encryption or malware guarantee. See actions and migration controls.

Bitwarden: manual organization session and ordinary history

Bitwarden offers fourteen ordinary organization Management API actions. This connects to the Public API, rather than the Vault CLI or Secrets Manager. Use an existing organization bearer session supplied by an authorized operator. The original node's client-ID/client-secret token acquisition is unsupported; this is an explicit, narrower migration.

Save a bitwardenOrganizationBearerSessionV1 account with an exact public HTTPS origin on port 443, organization UUID, session nonce, existing token and its actual UTC expiry. US cloud uses https://api.bitwarden.com; EU and public self-host require a separate regional acknowledgment. Self-host also requires an explicit server version. Identity origins are not contacted. Private/on-premise and IPv6-only hosts are unsupported by this transport. Token authority, actual expiry, organization identity, scopes, edition, entitlement, licensing and server compatibility remain unverified.

Named fields and Advanced JSON use the same strict encrypted, write-only account. Save/update/reopen is local and unverified with zero provider HTTP/DNS. There is no Test, login, token POST, refresh, discovery, identity lookup or schema fetch; bulk tests skip the profile. A replacement account invalidates earlier consent. See exact profile fields.

Names, emails, memberships, event IPs and acting-user identifiers can remain in ordinary execution/checkpoint/Human Task history and flow to later mapping, routing or AI after acknowledgment. Account encryption does not add raw-history encryption, automatic expiry, a scanner or deletion guarantees. Deletes and exact membership replacements may remove organization access. Review each resolved request separately. See actions and consequential effects.

Cortex existing key and n8n software Public API accounts

Cortex offers Get job (job.get) and Get job report (job.report) only. Save cortexExistingApiKeyV1 with an existing apiKey, exact public HTTPS443 origin and explicit basePath (empty only for an origin-root installation). The fixed /api/job path is appended once. Verify the instance version and read/analyze/orgAdmin permissions separately. Analyzer/responder execution and waitreport are unavailable.

n8n software Public API is separate from workflow import support. Save n8nPublicApiKeyV1 with an existing apiKey, exact public HTTPS443 origin and explicit configured API-root basePath, for example /api/v1 or /team/api/v1. Edition/version, API availability, scopes and entitlement remain unverified, including free-trial restrictions. The legacy n8nApi profile is an explicit migration, not an interchangeable account.

Named fields and Advanced JSON save the same encrypted, write-only account locally. Save/update/reopen sends zero HTTP/DNS and leaves authentication unverified. There is no Test, ping, schema fetch, discovery, OAuth, refresh or key acquisition; bulk credential tests skip these profiles. Select the exact saved account on each block. Saving a connection grants no request approval. See actions and controls.

The original run actor must confirm the exact resolved request with current workspace AND organization execution access. Selected encrypted account revision, published node/version, run input, live-policy generation, original task ID AND wait token and execution lease remain bound through one winning durable claim, settlement and fresh authenticated output. A replaced task never renews consent. One physical attempt is allowed, including workflow maxTries3; timeout, reset, rate limit or an uncertain used claim never triggers automatic resend, readback, redirect or refresh.

Names, IDs, tags, schema option labels, job/report fields and audit locations may contain personal data retained in ordinary run, checkpoint, Human Task history and downstream mappings or AI. The key uses existing credential encryption; this adds no raw-database encryption, automatic deletion/TTL, scanner or deletion guarantee. Local bounds are 64 KiB input, 1 MiB complete response, depth 32, 20,000 nodes, 4096 reads and thirty seconds for the whole invocation, rather than vendor guarantees. Real provider authentication, selected-instance permissions/schema/version/entitlement, durable DB/Redis recovery, native DNS/TLS/sockets, Review tutorials and real AI request interpretation remain untested; local production-owner tests use simulated dependencies.

n8n destructive Public API actions

The selected n8n Public API subset now has nine ordinary actions. Delete a credential (credential.delete, n8n_public_api_credential_delete), Delete an execution (execution.delete, n8n_public_api_execution_delete) and Delete a workflow (workflow.delete, n8n_public_api_workflow_delete) each send one destructive WRITE: DELETE /credentials/<credentialId>, /executions/<executionId> or /workflows/<workflowId>, prefixed by the saved API root. There is no query or body. Credential/workflow IDs are explicit 1–256 ASCII letters, digits, underscore or hyphen; execution IDs are positive safe decimal strings. No collection, wildcard or returned URL becomes a deletion target.

The existing encrypted manual API-key profile and unverified instance edition/version/scopes/entitlement/free-trial prerequisites still apply. All three require the initially unset schema-2 migration, raw-retention and one-attempt controls plus destructiveAcknowledged:true, then a separate original-actor Human Task showing the saved account, origin/root, exact DELETE path and named resource ID. Imported string "true", 1, null and invalid objects remain visible and unchecked until explicitly repaired; opening or saving another field never grants consent. Current workspace and organization authority, policy/account revision, original taskId and waitTokenId, immutable request and lease/claim/CAS checks still apply, including later authenticated OUTPUT.

A genuinely empty HTTP204 or HTTP200 response maps to data:{resource,id,response:"empty"}, with zero examined/returned records and outcomeVerified:false; it observes a response and does not prove deletion. HTTP200 expected metadata maps to response:"metadata" and bounded metadata, with an exactly matching target ID. Credential metadata is only id/name/type; execution data omits runData/workflowData/customData; workflow data omits parameters/settings/pinnedData/connections/credential references. JSON null, whitespace, malformed/unexpected bodies, wrong IDs and other statuses are withheld. There is no deleted:true, automatic readback, paging, polling, retry or uncertain-effect replay. Map node.<id>.json.data.id, .response or available .metadata.name/.status; names and IDs may contain PII retained in ordinary history and downstream mappings, without a new raw-database encryption, TTL/deletion or scanner guarantee.

Real deletion effects, provider authentication, durable DB/Redis, native DNS/TLS and matching-revision Review remain untested. Use only separately authorized disposable resources for live verification; existing credentials, workflows and executions are not test targets. Credential create, workflow create/update/activate/deactivate/getVersion and Tool/Trigger remain unavailable; nine actions do not establish full-node parity.

AWS Transcribe job requests and metadata

The partial AWS Transcribe subset supports four ordinary numeric-version-1 actions: transcribe.transcriptionJob.create, transcribe.transcriptionJob.delete, transcribe.transcriptionJob.get and transcribe.transcriptionJob.getAll. Transcript downloading, simplification, automatic language detection, unbounded pagination and Tool/Trigger forms are unsupported. This is partial service coverage, not full-node or provider parity.

Save an existing awsTranscribeManualSigV4V1 account with profile aws-transcribe-manual-sigv4-v1, partition aws, one of twenty pinned standard regions, declared twelve-digit account ID and manual access keys. Named fields and Advanced JSON use the same encrypted, write-only account. manual-static omits token and expiry; manual-temporary requires an explicitly supplied session token and actual UTC expiry, with more than ninety seconds remaining at execution. Save/update/reopen is local and unverified with zero provider HTTP/DNS. Test and bulk Test are unavailable. No ambient/system/role credentials, STS acquisition, refresh, discovery or custom endpoints are used. The declared account does not prove AWS identity.

Import exact n8n-nodes-base.awsTranscribe with numeric typeVersion:1, preserve original parameters and credential references, and repair the schema-2 input deliberately. Select the exact saved account and set the initially unset migrationAcknowledged:true and rawRetentionAcknowledged:true controls. Imported strings such as "true", numbers and null remain invalid, unchecked and preserved until an explicit click or repair. Unknown and inactive fields remain repair items; embedded credentials are refused before source-ledger persistence. Original get defaults to returnTranscript:true; that intent is unsupported and is never silently converted to metadata.

Create requires an explicit unique transcriptionJobName, authorized same-region s3://bucket/key media URI, mediaFormat, detectLanguage:false, supported languageCode and an explicit options object. The original v1 node has no media-format field, so imports require manual format selection. Supported original settings include channel identification, maximum alternatives/speaker labels (2–10), vocabulary name and a complete vocabulary-filter name/method pair. Channel identification cannot combine with speaker labels. Deliberately set the initially unset processingAcknowledged:true: AWS obtains the authorized media, processes speech, may charge costs and retains job/transcript data under its policies; the app never fetches the media URI. A QUEUED or IN_PROGRESS result remains pending. No polling or completed-transcription claim follows submission.

Delete requires the explicit job name and initially unset destructiveAcknowledged:true, then a separate original-actor Human Task showing the exact AWS request. An actual empty HTTP200 body yields empty operationData:{} with responseReceived:true and outcomeVerified:false. It is response evidence only; job deletion completion and transcript object removal are not verified. There is no fabricated success:true, hidden readback or automatic resend.

Get requires returnTranscript:false explicitly. It returns bounded typed job/status/time/media/transcript-location metadata in operationData. Transcript locations remain inert JSON strings, including on COMPLETED jobs: the app never fetches a provider-returned URI or simplifies a transcript. Get All requires explicit returnAll:false, limit 1–100, filters:{} or supported original status/name filters, and boundedPageAcknowledged:true. It returns one partial page envelope with TranscriptionJobSummaries and any actual top-level NextToken. Copy that token into nextToken only for a separate manually approved next request. A short or empty page does not prove exhaustive inventory; the original broken unbounded paginator is not reproduced.

Only the original run actor can approve the exact request within fifteen minutes. Current workspace and organization authority, selected account, published node/version, run input, resolved request, original task ID, wait token, stable durable decision and execution lease are checked through claim, dispatch, settlement and fresh OUTPUT. Replacement never renews consent. One claim permits at most one physical attempt despite workflow retries; timeout, revocation, disconnect, expired lease or uncertain settlement cannot authorize resend. Same-lineage failure cleanup grants no new authority and remains fenced by the real lease.

Local limits are 16 KiB request, 1 MiB whole response, thirty seconds, depth32/nodes20000 and strict unknown-field refusal. Malformed, oversized, unexpected, numerically mistyped or credential-reflecting responses fail closed. Useful job metadata, names and media/transcript locations enter ordinary workflow/checkpoint/Human history and may reach downstream AI. Accepted raw response bytes use the existing encrypted protected Workspace Files mechanism; this adds no raw-database encryption, automatic TTL, plaintext-scanner or deletion guarantee.

En español: selecciona una cuenta manual existente y la acción exacta. Corrige los campos importados y marca cada confirmación deliberadamente. AWS obtiene el archivo autorizado; la aplicación no descarga archivos ni transcripciones. Los trabajos pendientes siguen pendientes, eliminar solo acredita la respuesta y listar devuelve una página parcial con un token manual. Los metadatos y sus ubicaciones pueden conservarse en el historial y llegar a la IA posterior.

Local tests simulate provider/Prisma facts and use actual public resume, Human Task, checkpoint and output owners. Real AWS authentication/permissions/effects, native DNS/TLS, DB/Redis restart, real AI request interpretation and Review workspace/organization Help walkthroughs remain untested because those external environments and credentials are outside this worker's authority. See the official StartTranscriptionJob, DeleteTranscriptionJob, GetTranscriptionJob and ListTranscriptionJobs contracts.

AWS Certificate Manager certificates

The five ordinary numeric-version-1 actions are acm.certificate.delete, acm.certificate.get, acm.certificate.getMany, acm.certificate.getMetadata and acm.certificate.renew. Select AWS Certificate Manager, then the exact certificate action. Tool/Trigger variants, certificate requests/import/export, private-key export and ACME management are unavailable.

Use an encrypted saved awsAcmManualSigV4V1 account with profile aws-acm-manual-sigv4-v1, partition aws, a supported pinned commercial region, declared twelve-digit account ID, access key ID and secret access key. Named fields and Advanced JSON save the same account. manual-static omits session token/expiry; manual-temporary requires the actual supplied session token and UTC expiry; execution requires more than ninety seconds remaining. Local Save can retain an expired profile as unverified, but execution refuses it without refresh. The current bounded endpoint profile supports the same twelve explicitly pinned regions as Comprehend; other regions and partitions require a separate profile. The declared account ID restricts the certificate ARN and is not verified identity. Save/update/reopen is local, encrypted, write-only and unverified, with zero provider HTTP/DNS. Test and bulk Test are unavailable. No automatic STS, AssumeRole, system credentials, refresh, custom/FIPS/dualstack endpoints or discovery occurs.

Import n8n-nodes-base.awsCertificateManager with numeric typeVersion:1, preserve its original node/parameters/credential references, then reselect the native account and repair schema 2 explicitly. Unknown and inactive values stay visible and refuse execution. Embedded credential material is refused before source-ledger/workflow persistence. Original authentication:iam maps to the manual account lifecycle; AssumeRole/system/custom endpoint settings require manual repair. Remove the original unused S3 bucketName and certificateKey fields explicitly; they never become ACM request fields. The original default renew is not consent. Portable imports and exports can retain non-secret domain/certificate business data and original references.

Set parameterSchemaVersion:2, migrationAcknowledged:true and rawRetentionAcknowledged:true. Acknowledgments start unset, require literal boolean true and retain imported string "true" as invalid and unchecked until explicitly repaired. Delete also requires destructiveAcknowledged:true; renew requires providerEffectsAcknowledged:true. The exact certificate ARN must match the selected region/partition/account. Only the original run actor can approve the exact named request within fifteen minutes, with current workspace AND organization authority. The runtime captures the original task ID, wait token and durable decision at resume production, then loads that original task by ID throughout claim, send, settlement and fresh OUTPUT. A replacement task, wait, decision, account, node or input cannot renew consent. The execution lease remains required; an expired lease cannot weaken the physical-claim or cleanup fence.

Delete removes the certificate and associated private key; an in-use certificate cannot be deleted. AWS documents eventual consistency and unchanged Private CA billing. Renew requests renewal of an eligible certificate and can require Private CA permissions and incur costs. Both are writes: an empty HTTP 200 proves only that the response was observed, and the result retains outcomeVerified:false. There is no completed-deletion/renewal claim, readback, polling or certificate acquisition. Get returns only valid certificate PEM blocks and optional certificate chain. Get Metadata returns useful bounded exact certificate/domain/subject/contact/status/time/algorithm/type data. Those fields remain available for downstream mapping.

Get Many reads exactly one page, with original limit 1–500/default100, returnAll:false and boundedPageAcknowledged:true. Preserve the nested original options.certificateStatuses, options.extendedKeyUsage, options.keyTypes and options.keyUsage arrays. Extended Key Usage applies independently of status, correcting the original gate. Omit unused filters; explicit empty arrays require repair. AWS defaults restrict types to RSA_2048 and omit ACME origins, so a short/empty page does not prove exhaustive account inventory. NextToken remains returned business data and is not followed. Return All, continuation input and newer sort/origin/export filters require separate explicit extensions.

Local bounds are 16 KiB request, 1 MiB whole response, thirty seconds, depth32/nodes20000 and one physical attempt, even when workflow retries are configured. AWS permits a GetCertificate chain up to 2 MiB; this application's narrower whole-response bound refuses a valid oversized chain without truncation or fabricated success. Malformed/duplicate/invalid JSON, unexpected fields/status, credential reflections, private/reserved origins and partial responses fail closed. Once claimed, a timeout, disconnect, parser/storage failure or uncertain outcome never permits replay through a new gateway.

Exact accepted business output, inputs, domain names, ARNs, subjects, validation contacts and certificate PEM may remain in ordinary run/checkpoint/Human history and reach downstream AI or remote actions. Accepted raw response bytes are application-encrypted into protected Workspace Files and authenticated at fresh OUTPUT; the configured scanner sees ciphertext. This supplies no raw database encryption, automatic TTL, plaintext scanner or deletion-completion guarantee. Invalid responses and credential/error reflections do not become accepted business output.

For example, one partial list uses:

{
  "parameterSchemaVersion": 2,
  "migrationAcknowledged": true,
  "rawRetentionAcknowledged": true,
  "returnAll": false,
  "limit": 100,
  "boundedPageAcknowledged": true,
  "options": { "extendedKeyUsage": ["TLS_WEB_SERVER_AUTHENTICATION"], "keyTypes": ["RSA_2048"] }
}

An explicitly acknowledged deletion names its certificate:

{
  "parameterSchemaVersion": 2,
  "migrationAcknowledged": true,
  "rawRetentionAcknowledged": true,
  "certificateArn": "arn:aws:acm:us-east-1:123456789012:certificate/12345678-1234-1234-1234-123456789012",
  "destructiveAcknowledged": true
}

En español: elige la acción exacta y una cuenta manual guardada, corrige los campos importados y marca cada confirmación deliberadamente. El actor original confirma el ARN y la solicitud; HTTP 200 solo demuestra una respuesta, no la eliminación ni la renovación finalizadas. Una lista lee una página parcial y los datos del certificado permanecen disponibles para pasos posteriores con la retención habitual.

Local tests use simulated provider/store boundaries. Real AWS permissions/eligibility/effects, native DNS/TLS/provider execution, durable DB/Redis restart, storage scanning and Review workspace/organization Help walkthroughs remain untested. See the official DeleteCertificate, GetCertificate, ListCertificates, DescribeCertificate and RenewCertificate contracts.

DynamoDB typed items

Select AWS DynamoDB for the four ordinary numeric-version-1 operations dynamodb.item.upsert, dynamodb.item.delete, dynamodb.item.get and dynamodb.item.getAll. These use PutItem, DeleteItem, GetItem and an explicit Query or Scan. This is a partial service integration; automatic mapping, simplified decoding, discovery, Tool/Trigger variants and other APIs are unavailable.

Save awsDynamoDbManualSigV4V1 with profile aws-dynamodb-manual-sigv4-v1, partition aws, one of twelve pinned commercial regions, a declared twelve-digit account, and manual static keys or the supplied temporary session token with its actual UTC expiry. Temporary execution requires more than ninety seconds remaining. Named fields and Advanced JSON save/update/reopen the same encrypted write-only, locally unverified account with zero provider HTTP/DNS. Test and bulk Test are unavailable; no SDK, environment/metadata discovery, role assumption, STS acquisition, refresh or custom/FIPS/dualstack endpoint runs. A declared account is not verified AWS identity.

Import exact n8n-nodes-base.awsDynamoDb with numeric typeVersion:1. Original parameters, raw JSON types and credential references remain retained through edit, serialize, reopen and reimport. Reselect the typed saved account and deliberately repair parameterSchemaVersion:2, typed item/key/expression fields and unsupported legacy options. Embedded credential controls are refused before source-ledger persistence; ordinary typed database attribute names remain business data. The initially unset migrationAcknowledged, rawRetentionAcknowledged and onePhysicalAttemptAcknowledged must each be literal boolean true. Imported "true", 1 or null is never consent.

PutItem is full replacement, including removal of omitted attributes on an existing item with the same key. Supply the complete typed item and explicit replaceAcknowledged:true. Delete requires an exact typed key and deleteAcknowledged:true; choose returnValues:NONE or ALL_OLD. Both also require a separate original-actor Human Task for the exact table/item or key, conditions, account and consequences. Conditions can fail without retry or hidden readback. acknowledged:true, submitted input and returned old attributes observe the response; remotePersistenceVerified:false and outer outcomeVerified:false remain explicit. They never prove remote persistence.

AttributeValue JSON preserves S, exact decimal-string N (up to38 significant digits within the documented range), strict padded base64 B, BOOL, NULL:true, nonempty unique SS/NS/BS, L and M, including empty lists/maps. Primary keys use nonempty scalar S/N/B values. Attribute names and table schema are entered manually, without discovery. Set/list and data-name bounds may be narrower than AWS. GetItem allows explicit consistency and projection aliases; a missing Item returns found:false, while malformed or unexpected bodies fail closed.

Get All requires an explicit mode:Query with keyConditionExpression, or mode:Scan; index, filter, projection, aliases and typed values are manual. Set returnAll:false and limit:1..100: Limit counts evaluated items, not returned items or a cost guarantee. Exactly one bounded page is read. An empty filtered Items array can still return a typed lastEvaluatedKey; copy it to exclusiveStartKey only in a new, separately confirmed request. select:COUNT returns a count/scanned-count object, never an array. COUNT cannot include a projection. pageComplete means this response had no continuation key, not snapshot or exhaustive-inventory proof.

Local whole-input/request bounds are64KiB, whole response1MiB, depth32,20000nodes,4096reads and thirty seconds. Inherited32KiB signed approval and48KiB Human instructions can impose a smaller effective input. Oversized, missing, malformed, duplicate or unexpected bodies are refused without truncation. One durable physical claim remains bound to the actual public resume producer's original task ID, wait token and durable decision, current workspace and organization authority, account/input revision and live acquired lease through settlement and fresh authenticated OUTPUT. Expired ownership or an uncertain used claim cannot replay through a new gateway; unchanged-lineage failure cleanup uses the execution-lease CAS fence.

Table names, keys, attribute names and item data may be PII retained in ordinary workflow/checkpoint/Human history and downstream mapping or AI. Account encryption adds no raw-database encryption, automatic TTL/deletion or scanner guarantee. Local tests simulate provider/store facts; live AWS, native DNS/TLS, real DB/Redis restart, real AI phrasing and matching-revision Review workspace/organization Help walkthroughs remain untested. See official GetItem, DeleteItem and AttributeValue contracts.

AWS ELB ordinary actions

Choose AWS ELB and the encrypted write-only awsElbManualSigV4V1 manual profile. Named fields and Advanced JSON save/update/reopen locally with zero HTTP/DNS and remain unverified. No Test, STS, role/system/metadata credentials, refresh or hidden discovery occurs. Seven ordinary actions are partial service coverage; use explicit ARN/network/cost/privacy consent and a separate original-actor Human Task. See actions, controls, limits and repair policy. Real AWS effects, native transport, durable DB/Redis and matching Review tutorials remain untested.

AWS Cognito User Pools admin

This preview implements thirteen ordinary numeric-version-1 admin actions: group create/delete/get/getAll/update, user addToGroup/create/delete/get/getAll/removeFromGroup/update, and userPool get. It is partial service coverage, not an OAuth or authentication/login provider, Tool/Trigger support or full node parity.

Save an encrypted write-only awsCognitoManualSigV4V1 account with profile aws-cognito-manual-sigv4-v1, partition aws, declared twelve-digit account ID and region us-east-1, us-east-2 or us-west-2. The pinned endpoint is that region's cognito-idp HTTPS443 origin. Choose manual-static keys or manual-temporary supplied token with its actual UTC expiresAt; dispatch requires more than ninety seconds remaining. Save/update/reopen via named fields or Advanced JSON is local and unverified, with zero provider HTTP/DNS. Test and bulk Test, ambient/system/metadata/role credentials, STS acquisition, refresh, discovery and alternate endpoints are unavailable. A declared account is not verified identity.

Set parameterSchemaVersion:2, migrationAcknowledged:true and rawRetentionAcknowledged:true. Acknowledgments start unset and require literal boolean true; imported string "true", 1, null and objects remain invalid, visible and unchecked until explicitly repaired. Use literal userPoolId, username and groupName. The original user locator can mean a sub requiring lookup: it never silently becomes Username. Original userPool, user, group, newUserName, newGroupName, simple, includeUsers, additionalFields and nested attribute forms remain manual repair items. There is no hidden DescribeUserPool preflight, sub lookup, ListUsersInGroup or dynamic loader.

Every write requires writeAcknowledged:true. User deletion targets the named identity; group deletion never automatically deletes users. Membership, RoleArn and precedence changes can affect token claims and access, while existing session effects remain unverified. Group create/update uses roleArn sent as AWS RoleArn, correcting the original Arn routing; precedence 0 is preserved. User create/update accepts userAttributes:[{"Name":"custom:team","Value":"Exact value"}]. Supply the exact AWS name, with custom: exactly once; duplicate names are refused, and values including empty text are never trimmed or flattened over user metadata. The response exposes useful bounded users, groups and pool metadata under operationData; exact accepted body bytes are application-encrypted in scoped WorkspaceFiles. Ordinary business data and signed Human Task input remain in history and can reach downstream mapping/AI; no raw-database encryption, TTL or plaintext-scanning claim is made. Unknown/credential/password echo fields, duplicate JSON, malformed responses and oversize data fail before accepted-artifact retention.

AdminCreateUser requires explicit messageAction SUPPRESS or RESEND, desiredDeliveryMediums, forceAliasCreation, credentialCreationAcknowledged:true and costAcknowledged:true. Cognito may create credentials and generate a temporary password. RESEND additionally requires messagesAcknowledged:true and can email/SMS, reset invitation expiry and incur charges; omit that acknowledgment with SUPPRESS. Force Alias Creation true additionally requires aliasMigrationAcknowledged:true and can move a verified email/phone alias from another user; omit it with false. Supply the selected delivery attribute explicitly. Caller-supplied temporaryPasswordOptions, TemporaryPassword, passwords and embedded account secrets are refused before workflow/source-ledger persistence; use a separately authorized credential workflow for those unsupported inputs.

Lists require returnAll:false, boundedPageAcknowledged:true and limit 1–60/default20. One request returns one partial page: groups use nextToken → NextToken; users use paginationToken → PaginationToken, with optional exact filter. Returned opaque cursor bytes are data for a later independently approved request. Empty/short pages do not prove completeness. There is no automatic pagination, retry, polling, readback, redirect or returned-URL fetch.

The original run actor must approve the exact request in a separate Human Task within fifteen minutes. The public resume producer captures immutable task ID, wait token and stable decision revision; current workspace AND organization access, account/input/node/policy/request and execution lease are checked through claim, send, response, settlement and fresh OUTPUT. One actual claim CAS permits one attempt. Replacement lineage, revoked authority, expired lease or uncertain/partial effects never grant replay or automatic cascade. Blank HTTP200 responses record responseReceived:true and outcomeVerified:false, never completed deletion/membership/update. Local limits: 16KiB request, 1MiB response, thirty seconds, depth32/nodes20000. Actual deployed entitlement, source-version equivalence, permissions, real AWS and Review workspace/organization walkthroughs remain unverified by mocked local checks.

AWS IAM identity and membership actions

The ordinary numeric-v1 IAM user/group subset covers twelve actions only. It uses the dedicated awsIamManualSigV4V1 account: profile aws-iam-manual-sigv4-v1, partition aws, signing region us-east-1, and fixed global https://iam.amazonaws.com/ over HTTPS443. Use named fields or Advanced JSON for manual static keys, or a supplied temporary token with its actual UTC expiry. Save and reopen are encrypted, write-only, local and unverified; Test is disabled. No ambient, system, role, metadata, STS, credential acquisition or refresh is used.

Exactly one requested Action is sent by form POST with Version 2010-05-08. Identity, access and membership changes require explicit acknowledgments and a separate original-actor Human confirmation. DeleteUser/DeleteGroup fail on provider dependencies; they never remove memberships, login profiles, MFA, policies or access keys implicitly. Membership add/remove are separate exact user/group operations. CreateUser supports path, tags and permissions boundary; updates support path and rename only. No policy, role, access-key or login-password CRUD is added. HTTP200 is provider response evidence, never verified effect completion.

Lists require explicit returnAll: false, limit 1–1000 and bounded-page consent. GetGroup also pages Users; IsTruncated and opaque exact Marker remain visible even when Users are omitted. Continue only with another separately approved request. No automatic paging, membership scans, discovery, loaders, post-read, retries, returned-URL fetching or redirects. Original helper, returnAll and includeUsers semantics that cannot be preserved remain visible for manual repair; unknown fields and invalid consent types survive export/reopen. Embedded credentials are refused before source-ledger persistence.

Typed user/group identity, ARN, path, creation time, tags, boundary and explicitly selected membership data plus an exact raw reference enter ordinary run/checkpoint/Human history and may reach downstream AI. Account encryption adds no raw database encryption, automatic TTL or scanner guarantee. A bounded XML decoder rejects DTD, external/custom entities, duplicate or wrong envelopes, malformed members, unknown fields/status and credential reflection before retaining raw data. Current workspace and organization permissions, account, input, policy, immutable original task/wait/decision and lease are rechecked through claim, send, response, settlement and fresh OUTPUT. Uncertain or expired claims cannot resend. Local bounds: 32 KiB request, 1 MiB response, 30 seconds. Real AWS, hosted Review and full node/provider parity remain untested.

AWS S3 ordinary bucket subset

The partial ordinary S3 subset covers only s3.bucket.create, s3.bucket.delete, s3.bucket.getAll and s3.bucket.search: four of twelve actions, with exact numeric source V1/V2 imports analyzed separately. Search is ListObjectsV2 object metadata, not a bucket-name search. Object upload/download/copy/delete, folder deletion, multipart, Tool and Trigger remain unsupported.

Save awsS3BucketManualSigV4V1 / aws-s3-manual-sigv4-v1 with partition aws, an explicitly known us-east-1 or eu-west-1 region, declared twelve-digit account ID and manual static keys or supplied temporary token plus actual UTC expiry. Temporary execution needs over ninety seconds remaining. Named fields and Advanced JSON use encrypted, write-only, unverified local Save/update/reopen with zero provider HTTP/DNS. Test, bulk Test, STS, system/role chains, refresh, custom endpoints and discovery are unavailable.

Every request requires schema2 and initially unset migrationAcknowledged, rawRetentionAcknowledged, regionAcknowledged and billingAcknowledged, all literal true, plus action-specific consent. Raw string "true", 1, null and objects remain visible and unchecked until a deliberate click. Then a separate original-actor Human Task approves the exact request. Current workspace and organization access, account, node, input, policy, original task ID/wait token/stable decision and real execution lease are checked through the one durable claim, send, response, settlement and fresh OUTPUT. Expired or uncertain claims never replay.

Create requires name, additionalFields:{} (or a matching known region) and name/global-namespace/security/write acknowledgments. It sends one PUT: empty body for us-east-1, matching XML LocationConstraint elsewhere. Only provider secure defaults are supported; all original ACL/grant/ownership/tag/object-lock intent stays for explicit repair. HTTP200 empty body yields response evidence with effectConfirmed:false; Location remains inert manual metadata. Creating an existing owned us-east-1 bucket can reset ACLs under AWS behavior; inspect the exact name and permissions manually before approval.

Delete requires name, empty-bucket/global-namespace/destructive acknowledgments and sends one DELETE. The bucket must already be empty of objects, versions and delete markers; nothing enumerates or deletes them or changes protection/ownership. Only HTTP204 empty body is accepted as response evidence, never confirmed deletion. A deleted global name may be reused by another account.

ListBuckets (getAll) needs explicit returnAll:false, limit:1..1000, filters:{} and boundedPageAcknowledged:true. This is a local cap, not the AWS maximum10000. Optional filters.bucketRegion must equal the endpoint region; filters.prefix filters bucket names. Output preserves actual ContinuationToken; copy it byte-for-byte into nextToken for a separately approved request. Unpaginated accounts with approved quota above10000 may be rejected; no auto-page or local slice is used.

ListObjectsV2 (search) needs bucketName, explicit returnAll:false, limit:1..1000, additionalFields:{}, bounded-page and Requester Pays acknowledgment. Supported typed options are prefix, delimiter, encodingType:url, fetchOwner, startAfter and requesterPays; the payer is a header. Dotted names use regional HTTPS443 path style. Output keeps bounded Contents, CommonPrefixes, KeyCount, IsTruncated, keys, ETags, sizes, storage classes, dates and actual NextContinuationToken; tokens and keys remain inert values. No content or returned URL is fetched. Empty XML containers and singleton/many entries normalize without fabricated rows.

Original versions keep exact parameters and credential references. V1 ignores the location region during signing; V2 forwards it and handles dotted search names differently. Both original location reads, null/us-east-1/EU assumptions, returnAll/default/limit and unsupported options require visible manual repair. No hidden GetBucketLocation/HeadBucket/readback, redirect, follow-up, polling or retry occurs. Local32KiB request/1MiB response/thirty-second bounds apply; malformed, duplicate, DTD/entity/XXE, wrong-envelope, oversized or credential-reflecting replies fail before accepted artifact retention. Useful metadata and tokens enter ordinary run/Human history and downstream AI; account encryption adds no raw DB encryption, TTL or scanner guarantee.

En español: selecciona una cuenta manual existente y una región conocida, corrige el origen V1/V2 y marca cada confirmación de forma deliberada. Crear y eliminar solo acreditan respuestas; listar devuelve una página parcial y buscar lista metadatos de objetos. Copia el token exacto para otra solicitud aprobada. No se descargan objetos ni se eliminan sus versiones.

Local tests simulate provider/Prisma facts. Real AWS credentials, permissions/effects, native DNS/TLS, DB/Redis restart, AI request interpretation and Review workspace/organization Help walkthroughs remain untested because they are outside this worker's authority. See CreateBucket, DeleteBucket, ListBuckets and ListObjectsV2.

Venafi certificate read accounts

Venafi TLS Protect Cloud and Datacenter use separate saved credential contracts. Both offer Named fields and Advanced JSON, encrypted write-only Save/update/reopen, and saved · unverified status with zero provider HTTP or DNS. Test is unavailable and bulk credential checks skip them.

• Cloud uses venafiCloudUserKeyV1: profile, exact region: "cloud" (US) or "eu", and an existing user apiKey. Requests send tppl-api-key to the corresponding fixed https://api.venafi.cloud or https://api.venafi.eu origin. A service-account access token uses a different Bearer contract and is unsupported here.
• Datacenter uses venafiDatacenterManualBearerV1: profile, an explicitly supplied canonical public HTTPS443 origin, accessToken, exact tokenScope: "certificate", actual tokenExpiresAt as UTC ISO8601 ending in Z, tlsVerification: true and manualLifecycleAcknowledged: true. The token must have more than 30 seconds left. Certificate scope plus View and Read permissions are prerequisites to verify separately. This manually supplied bearer differs from the original username/password/client-ID preAuthentication lifecycle; no password grant, token acquisition or refresh occurs.

Do not put a path, trailing slash, userinfo, query or fragment in the Datacenter origin. TLS verification stays enabled. The current outbound policy requires public DNS/IP results and IPv4; private-network and IPv6-only installations are unsupported by this adapter. Server version, trust configuration, permissions, regional availability and deployment reachability remain unverified. Repair imported credential references deliberately; saving never verifies provider access.

Original actor approval, current workspace and organization authority, saved account revision, exact input/node/live policy, original task ID/wait token/stable decision and live execution lease are checked before claim/send, after the response, at settlement and on fresh OUTPUT. One physical claim permits at most one attempt; expired or uncertain claims never replay. Same-lineage cleanup cannot mutate a replacement owner.

Only ordinary numeric version-1 certificate/getMany Cloud and certificate/get Datacenter nodes are adapted. There is no Tool/Trigger parity, download, private-key export, issuance, renewal, deletion, authentication lifecycle, automatic paging, retry, redirect or returned-URL fetch. See the exact read controls.

Official Cloud API setup distinguishes user keys from service-account tokens. Datacenter 26.1 certificate GET documentation describes certificate scope and View/Read permissions. These references do not verify your account or deployment.

TheHive v1 manual case reads

TheHive and TheHiveProject are two imported wrappers of one Query API service. Only exact ordinary version-1 n8n-nodes-base.theHive and n8n-nodes-base.theHiveProject with explicit case/get are supported. Choose case.getLegacy with theHiveApi and explicit apiVersion: "v1", or case.getProject with theHiveProjectApi. Legacy input is { "id": "~case-id" }; Project input is { "caseId": "~case-id" }. Project also accepts the original manual ID locator. Tool/Trigger variants, missing versions, v0, writes, options, arbitrary queries, list/URL locators, loaders, binary and asynchronous actions require manual repair. Original source controls remain retained and checked.

Save an existing API key with an explicit public HTTPS origin (no subpath, credentials, query, fragment or custom port), exact wrapper profile and an explicit organization policy:

{
  "profile": "theHiveApi",
  "apiVersion": "v1",
  "url": "https://incident.example.org",
  "ApiKey": "SYNTHETIC_EXISTING_API_KEY",
  "organization": { "mode": "named", "name": "responders" },
  "allowUnauthorizedCerts": false
}

For Project, use profile: "theHiveProjectApi" and omit apiVersion. Named fields and Advanced JSON use the same validator; save, update and reopen are local and unverified, with zero provider HTTP. Test and bulk testing are unavailable. Choose { "mode": "default" } explicitly to use the API key owner's default organization, or named scope to send X-Organisation. The application does not discover or verify that organization. A private-only installation is outside this public-origin policy; deployment outbound policy and reachability remain unverified.

Both reads need a separate signed 15-minute Human Task decided by the original run actor. Current workspace and organization membership, account, workflow, input, policy, task, wait, durable decision and execution lease must still match. One physical claim permits one bounded POST to /api/v1/query?name=get-case-…; retries, polling, paging, refresh and returned-link fetching are unavailable even if node retries are enabled. Legacy sends only getCase; Project adds exactly page 0 to 10 with extraData: ["attachmentCount"].

Outputs retain the wrapper's sourceIdentity, case ID, case number, title, unknown business fields and Project attachment count. Zero results return kind: "no-record"; incomplete/malformed/error/mismatched responses are refused. Numeric tokens that cannot pass the explicit precision policy are refused instead of rounded. Only a fully validated bounded response may be retained as an application-encrypted Workspace File; the configured scanner sees ciphertext. Useful business data can persist in workflow history and downstream steps. Fresh OUTPUT admission verifies current authority and the exact signed receipt and raw artifact, with no resend after uncertainty. Installed API version, provider acceptance, key permissions, actual organization visibility, native network behavior and matching-revision Review remain unverified; this is partial support.

Odoo contact reads

The public preview odoo provider supports exactly two ordinary numeric version1 actions: contact.get and contact.getAll. Use the saved odooApi JSON account with exact root public HTTPS url, explicit db, username, and password (a password or existing RPC API key, at least eight characters). Named fields and Advanced JSON save/update encrypted write-only secrets locally with zero provider HTTP or DNS. Save remains unverified; Test and bulk Test are unavailable.

Every run requires a separate original-actor Human Task showing the exact origin, database, username, contact query, selected fields, offset and limit. One approval covers exactly two POST /jsonrpc requests: JSON-RPC2 common.login, then object.execute on res.partner with the validated returned UID. The two stages share one total thirty-second budget and at most two HTTP attempts. Each stage is durably claimed and fenced by the current workspace and organization authority, selected account revision, input/policy/node, original task/wait/decision and lease. An ambiguous login or read cannot be replayed by a fresh gateway.

contact.get requires one canonical positive safe decimal-text contactId and options.fieldsList, a nonempty array of up to32 unique field names. contact.getAll requires actual returnAll:false, numeric offset0–1000000, numeric limit1–100 and the same field selection. The original contact branch has no filters: imported filter/custom/unknown options remain visible for manual repair and are refused, never silently discarded. Imported type versions, node identity and original credential references stay in a source ledger; select the exact saved account explicitly. Tool/Trigger, other models, writes, fields_get, loaders, unbounded returnAll, token acquisition, discovery, retry and automatic paging are unavailable.

The useful typed result preserves bounded contact objects, source IDs and unknown provider fields. Empty results are distinct no-record responses. GetAll returns page.partial:true, manual continuation intent and a suggested offset derived only from the rows received; no total or hasMore is fabricated, and a short/empty page does not establish completeness. Unsafe numeric tokens, precision-losing decimal forms, invalid envelopes, mismatched response IDs, provider errors, credential reflection and oversized responses refuse before accepted raw retention. Exact validated stage responses are encrypted in bound workspace files; selected business contacts may still enter ordinary history and downstream AI. attempts:2 describes the completed two-request protocol, with outcomeVerified:false.

Official Odoo19 source retains common.login and its noninteractive RPC API-key authentication path. The legacy endpoint is deprecated with eventual Odoo22/Online21.1 removal, not currently retired by that notice. Deployment/version support, Custom plan entitlement, public reachability, model and ACL permissions remain unverified. Local tests use simulated provider and Prisma facts; real provider, native DNS/TLS, PostgreSQL/Redis and Review workspace/organization walkthroughs have not been performed.

Flow task/get partial compatibility

Only ordinary n8n-nodes-base.flow numeric typeVersion1 task/get is supported. This is partial compatibility derived from the shipped node and helper, with live protocol and account entitlement unverified. A bounded unauthenticated GET of the official developer documentation failed DNS ([Errno 8] nodename nor servname provided, or not known); that failure does not show provider retirement.

Save flowManualBearerV1 with named fields or Advanced JSON: {"profile":"flowManualBearerV1","organizationId":"701","accessToken":"YOUR_EXISTING_TOKEN"}. Enter secrets only in the encrypted account editor. Save/update/reopen is local, write-only and unverified with zero provider HTTP; Test is unsupported and bulk Test skips it. No OAuth, token discovery, acquisition or refresh occurs. Provider organizationId is independent of application organization/workspace IDs and never inferred from them.

Use task.get with parameterSchemaVersion:2, migrationAcknowledged:true, rawRetentionAcknowledged:true, onePhysicalAttemptAcknowledged:true and one exact positive decimal taskId. A safe integer input is accepted; unsafe numbers refuse rather than round. Decimal strings preserve large IDs. Optional filters.include accepts only schedule, files, file_associations, parent, at most once each. Example: {"parameterSchemaVersion":2,"migrationAcknowledged":true,"rawRetentionAcknowledged":true,"onePhysicalAttemptAcknowledged":true,"taskId":"9007199254740993123","filters":{"include":["files"]}}. Acknowledgments start unset and require explicit review. Unknown optional controls remain visible manual repair intent; imported source parameters, versions and credential references remain in a source ledger. Reselect the typed saved account deliberately.

The original actor with current workspace AND application organization authority separately approves exactly one GET https://api.getflow.com/v2/tasks/{taskId}?organization_id={credentialOrganizationId} with Bearer accessToken. Original task ID, wait token, stable decision revision, actor/account/input/node/policy and execution lease remain bound through one physical claim CAS, response, settlement and fresh OUTPUT. Ambiguous responses, changed ownership or expired leases never grant a resend, including with node Retry enabled. Local limits are thirty seconds, 1MiB response, depth32 and 20k nodes; these are not provider limits. Self-hosted outbound reachability remains unverified.

Use data.task and optional included entities from the intact response envelope. Unknown useful business fields remain available; every provider JSON number is represented as its exact original text, including precision and exponent spelling. Original source IDs are not replaced by labels or application IDs. Accepted raw UTF-8 is application-encrypted in scoped WorkspaceFiles and represented by rawReference; plaintext raw/sourceItems are absent from ordinary output. Business data can remain in workflow history and downstream mappings/AI. The scanner sees ciphertext, not plaintext; no raw-database encryption, deletion deadline or backup deletion claim is made. Invalid envelopes, business errors, missing/mismatched tasks, auth/status/transport/rate failures, duplicate JSON, oversize bodies, invalid UTF-8 and secret reflection refuse success and accepted raw retention.

Create/update/getAll, triggers, webhooks, dynamic loaders, paging, Link helpers, retries, polling, readback and returned-URL fetch are unsupported. Mocked provider and Prisma facts do not verify live Flow, native DNS/TLS, durable DB/Redis, real AI or Review workspace/organization walkthroughs.

Splunk existing management bearer account

Choose Splunk and save a dedicated splunkExistingManagementBearerV2 account using Named fields or Advanced JSON. Supply the original management authToken and an explicit baseUrl, such as https://splunk.example.com:8089; canonical HTTPS443 is also accepted. A HEC token is not a management token. No origin path, userinfo, query, fragment, alternate port or TLS downgrade is accepted. TLS verification remains enabled. Original splunkApi references, allowUnauthorizedCerts:true and unknown credential options require explicit manual repair rather than silent conversion.

{
  "profile": "splunkExistingManagementBearerV2",
  "baseUrl": "https://splunk.example.com:8089",
  "authToken": "<existing-management-token>"
}

Save, update and reopen use encrypted, write-only local storage, remain unverified and send zero provider HTTP/DNS. Test is unavailable and bulk workflow Test skips this account. There is no token acquisition, refresh or scope discovery. Public outbound-address policy and pinned DNS/TLS apply; private and IPv6-only instances are unsupported by this transport. Deployment reachability and current provider acceptance remain unverified. Prior official-documentation 403/redirect responses establish neither acceptance nor retirement. See the exact metadata operation.

Azure Cosmos DB: get one existing container

The three acknowledgment switches start unchecked on a new or original n8n import. Choose each explicitly. Only literal boolean true counts as consent; imported strings such as "true", other invalid JSON values and expressions remain invalid through unrelated edits and export/reopen until explicitly repaired.

Select Azure Cosmos DB → Get container for the ordinary numeric n8n version 1 container/get subset. Save an existing account in Named fields or Advanced: edit JSON using exactly:

{
  "profile": "azureCosmosDbMasterKeyV1",
  "account": "your-public-account",
  "database": "ExactDatabaseId",
  "key": "YOUR_EXISTING_CANONICAL_BASE64_MASTER_KEY"
}

Account is the public Azure account name, without a URL. Database and container IDs preserve case and leading zeros; the initial supported ID alphabet is ASCII letters, digits, underscore, plus, equals, comma, dot, at sign and hyphen, 1–255 characters, excluding . and ... Each path segment is encoded once. The database belongs to the credential; the action supplies container. Prefer an existing read-only master key. Save/update/reopen encrypt locally, send zero HTTP and remain unverified; Test is unavailable and bulk Test skips the account. SQL/NoSQL API, local key auth, read-only-key rights, RU charges and network ACLs remain unverified. No Azure resource or identity is created.

Reselect the exact saved account and explicitly check schema-2 migration, normal-history retention and one-attempt acknowledgments. Enable WORKFLOW_LIVE_AZURE_COSMOS_DB=true independently of the current workspace Azure Cosmos DB live preference. The original actor needs current direct organization and workspace membership. One separate Human Task binds actor, organization, workflow/version/node, account ID/revision, origin, database, container and retention. Name matching or another account never substitutes for a deleted, rotated or unauthorized selected account. Approval expiry, cancellation, lease loss or any changed binding refuses the request.

The runtime makes one HMAC-SHA256 authenticated HTTPS GET to https://{account}.documents.azure.com/dbs/{database}/colls/{container}, with x-ms-date freshly generated at dispatch and x-ms-version: 2018-12-31. It signs the lower-case verb, resource type and date plus the case-sensitive encoded resource path, including the final blank line. Keys are decoded from canonical base64; no unsigned fallback exists. Public DNS and TLS are pinned and revalidated; private/custom/sovereign origins, alternate ports and redirects are refused. No item or partition-key probe, query, pagination, retries, management API or returned-link fetch occurs.

A successful 200 returns data.container, preserving exact string IDs, resource metadata and bounded unknown JSON business fields, including nested fields called token, password, __proto__ or constructor. Those values can contain private data and remain in ordinary history and downstream mappings. Raw transport text is discarded. Actual credential value echoes, duplicate JSON keys, unsafe integers, overflow, malformed UTF-8/JSON, mismatched ID, empty body and status errors refuse success. Limits are one total 30-second dispatch/read/decode budget, 1 MiB response, depth 32, 20,000 nodes and 4,096 reader steps. A durable claim and consumed-decision annotation are written together before HTTP; uncertainty or removed/tampered effects never authorize replay.

Original imports require exact ordinary numeric version 1, explicit container/get, By ID locator and simple:false. Original omitted container operation defaults to getAll and omitted simplify defaults true; unsupported controls outside the admitted listing profile require repair; source controls outside this original read contract, Tool/Trigger, extra auth/parameters and dynamic locators require visible manual repair. Source intent remains in the import ledger. OAuth, managed identity and connection strings are unsupported. Local simulations do not verify real Azure, native deployment, durable database/Redis restart or Review organization/tutorial behavior.

YouTube supplied video token account

Choose YouTube and save a dedicated youtubeVideoManualBearerV1 account in the current workspace. Named fields and Advanced JSON use the same central parser:

{ "profile": "youtubeVideoManualBearerV1", "accessToken": "YOUR_ALREADY_ISSUED_ACCESS_TOKEN" }

Save, update and reopen are encrypted, write-only and unverified, with zero provider calls. No Test connection, OAuth acquisition, refresh, scope discovery or Google Workspace credential substitution occurs. Obtain the token separately, enable YouTube Data API v3 in its Google project, and check expiry, quota and grants needed for private data. Application organization membership is separate from Google project/account identity; organization-scoped credential records are unsupported.

The selected saved account ID is binding: deleting, rotating, retyping or deauthorizing account A cannot silently use account B, even with the same name. See the one-video operation. Real provider access, quota, private-data grants and Review behavior remain untested.

X (Twitter v2): one username lookup

The useful native unit is exact ordinary numeric typeVersion: 2 user.searchUser, using one explicit username of 1–15 ASCII letters, digits or underscores. It sends one GET https://api.x.com/2/users/by/username/{username} with no query or body. The current documented hostname differs from original api.twitter.com; support for that old alias remains unverified, and redirects are refused. Original default tweet/resource conversion, v1/OAuth1, me/ID lookup, writes, extra fields, expansions, multiuser, paging and retries are visibly unsupported. Import retains original source ledgers and requires explicit repair and account reselection.

In Connections, choose X (Twitter v2) and enter your existing bearer in the named accessToken field, or Advanced JSON {"profile":"xUserV2ManualBearerV1","accessToken":"YOUR_EXISTING_TOKEN"}. Save/update/close/reopen use existing encrypted storage, remain unverified and send zero HTTP/DNS. Test is unsupported and bulk Test skips this account. OAuth2 PKCE acquisition/refresh and original OAuth or broad-scope aliases are not provided. Current X endpoint authorization and credits/pay-per-use require a suitable account; actual grants, entitlement, credits and visibility remain unverified without a real provider test.

In Builder, select user.searchUser, the exact saved workspace account and the username, and explicitly check migrationAcknowledged, outputRetentionAcknowledged and onePhysicalAttemptAcknowledged with schema version2. Current workspace live preference AND the operator WORKFLOW_LIVE_TWITTER gate are required. The original actor needs current organization AND direct workspace execution membership and confirms the exact material in a separate Human Task. Current account bytes, identity, original wait/decision, workflow/version/input, consumed resume segment and nonempty owned lease are checked before the physical claim, HTTP and output persistence. Rotation, deletion, retyping, deauthorization, membership/live changes, later resume/retry/recovery, removed/tampered claims and expired leases refuse further dispatch; uncertain outcomes never automatically replay.

The bounded original data user object is returned as data and one original-style item in sourceItemsJson. String snowflake data.id keeps precision. data.name, data.username and bounded unknown nested business fields are available to ordinary output, run history and downstream mappings, and may contain personal information. The raw response envelope and unknown outer fields are discarded; no plaintext raw artifact or new encryption purpose is introduced. Returned URLs remain inert. Explicit credential VALUES (including encoded forms) are refused on reflection; unrelated business field names are retained. Not-found, rate-limit, credits, auth, duplicate keys, malformed/unsafe values and partial errors fail instead of returning an empty user.

Local bounds are one physical attempt, total30s, response1MiB, depth32 and20k nodes. These are local bounds, not vendor quotas or a deletion/scanning guarantee. Current provider service, Review organization walkthrough, native DNS/TLS and durable DB/Redis restart are untested; local HTTP/Prisma/DOM simulations do not establish them.

PayPal: one payout item read

The native partial operation is ordinary n8n-nodes-base.payPal, numeric typeVersion: 1, payoutItem.get. It sends exactly one GET /v1/payments/payouts-item/{payoutItemId} on the saved explicit live (https://api-m.paypal.com:443) or sandbox (https://api-m.sandbox.paypal.com:443) environment. Enter one explicit string ID; numeric coercion, paths, query controls and returned links are refused.

Save a dedicated PayPal account through Named fields or Advanced JSON:

{
  "profile": "paypalPayoutItemManualBearerV1",
  "environment": "sandbox",
  "accessToken": "YOUR_CURRENT_MANUALLY_SUPPLIED_TOKEN"
}

Both editors use the same central validator. Save, update, close and reopen are local encrypted, write-only and unverified, with zero provider HTTP or DNS. Test is unavailable and bulk Test skips this profile. The original payPalApi client ID/secret credential and its hidden authorization POST are separate and unsupported. No OAuth acquisition or refresh occurs; replace an expired manual token yourself. sanbox, omitted environments and arbitrary fallbacks are refused. PayPal Payouts enablement, token expiry, merchant transaction visibility, current permissions and quota remain unknown without real account verification. The application's organization is distinct from the PayPal merchant identity.

Use parameterSchemaVersion: 2, an explicit payoutItemId, and independently check migrationAcknowledged, outputRetentionAcknowledged and onePhysicalAttemptAcknowledged. These acknowledgments start unset. Imported unsupported resources, versions, extra parameters and original credential references remain visible in the source ledger for manual repair and explicit account reselection. Only the original actor with current direct workspace and organization membership can approve the exact account, environment, route and retention in a separate Human Task. Current saved account material, live environment gate WORKFLOW_LIVE_PAYPAL, workspace preference, original task/wait/decision, resume segment and lease are rechecked at claim, dispatch, settlement and output admission. A removed or changed effect record, expired lease, deletion, rotation, retyping, deauthorization or lost membership cannot authorize a resend or account fallback.

The useful single payout-item object appears in result.data and the one-element result.items array. It retains string identifiers, nullable values and nested unknown business fields, including recipient and financial data, in ordinary execution history and downstream mappings. Review those fields before sending them to another action or AI step. The bounded raw HTTP text and transport envelope are discarded after parsing; no encrypted raw-response archive, new TTL, deletion or scanner guarantee is added. Exact credential values echoed in any key or value are refused; business fields are not removed because of their names. Provider errors and not-found responses are failures, while a payout item's documented business status such as FAILED is preserved as business data.

One consumed decision permits one physical request with a total real 30-second deadline, 1 MiB response, depth 32, 20,000-node and finite-read limits. Duplicate JSON keys, invalid UTF-8/media/status, numeric precision loss, overflow, redirects and ambiguous outcomes fail closed with no automatic retry. Payout create/get/list, payout-item cancel, other resources, triggers, dynamic loaders, pagination, returned-link fetching and token acquisition/refresh remain unsupported. This one selected read is not complete PayPal parity. Tests use simulated HTTP/Prisma/DOM facts; real PayPal, native DNS/TLS, a durable database restart, real AI interpretation and the matching authenticated Review workspace/organization tutorial walkthrough remain untested.

Primary endpoint contract: PayPal payout item details.

Google Cloud Firestore: supplied document token

Save a dedicated Google Cloud Firestore workspace account in Named fields or Advanced JSON, using exactly firestoreDocumentManualBearerV1:

{
  "profile": "firestoreDocumentManualBearerV1",
  "source": "manual",
  "tokenType": "Bearer",
  "accessToken": "YOUR_EXISTING_ACCESS_TOKEN",
  "expiresAtMs": 1800000000000,
  "scope": "https://www.googleapis.com/auth/datastore",
  "principalType": "unknown"
}

Replace the sample expiry with the token's claimed Unix-millisecond expiry. It must remain more than thirty seconds away at each check. principalType must explicitly be unknown, user or serviceAccount; these are supplied claims, not identity verification. The token's audience remains unknown. No Google Workspace credential, original googleApi service-account JSON or googleFirebaseCloudFirestoreOAuth2Api record is silently substituted. No OAuth acquisition/refresh, JWT signing, token exchange or scope discovery occurs.

Save, update, close and reopen use existing encrypted, write-only local account storage, send zero provider HTTP/DNS and remain unverified. Test is unavailable and bulk Test skips this profile. Enter secrets only in the account editor. Check Firestore API enablement, database mode, IAM versus Security Rules, actual token grants/expiry and quota separately. The application's organization is distinct from a Google project or principal; organization-scoped credential records are unsupported.

See the one-page typed document operation. Real Firestore, native deployment, durable database restart and authenticated Review organization/tutorial behavior remain untested; local simulated Prisma/HTTP/DOM checks do not certify them.

To replace a saved account, reopen it, re-enter the complete valid profile in Named fields or Advanced JSON and choose Update account. This updates the same account ID and keeps its replacement secret encrypted, write-only and unverified; reopening never reveals the saved token. Advanced JSON and Named fields share the exact profile validator.

The same Firestore token can be selected for six additional bounded native read/write actions. Writes require their own literal effect acknowledgment and exact Human approval; saving the token grants neither. See the action contracts and migration differences.

Microsoft Dynamics: one Dataverse account read

The native Dynamics profile supports account.get, account.getAll, account.create, account.update and account.delete on the saved public commercial HTTPS {environment}.crm.dynamics.com origin at port443, using /api/data/v9.2/accounts. Other regions, sovereign/on-premises endpoints, other tables, lookups, loaders and Tool/Trigger variants remain unsupported.

Save the dedicated dynamicsAccountManualAccessTokenV1 profile in Named fields or Advanced JSON with accessToken, exact environmentOrigin, matching declared audience and Unix-millisecond expiresAtMs. Save/update/reopen are encrypted, write-only, local and unverified with zero provider HTTP/DNS. Test and OAuth acquisition/refresh are unavailable; historical microsoftDynamicsOAuth2Api credentials are never converted. Declarations do not verify real token grants, tenant consent, licensing or security roles. At least30 seconds of declared lifetime must remain at execution.

Create/update accept the supported text, Int32, four-decimal Money, stage GUID and up to two address rows. Other columns, deep inserts and lookup bindings require another integration. Dataverse plug-ins, flows, business rules and auditing may run for writes. HTTP acceptance does not verify their completion; an uncertain write may already have occurred and never retries automatically.

Set parameterSchemaVersion:1, then independently select literal migrationAcknowledged, rawRetentionAcknowledged and onePhysicalAttemptAcknowledged. Create/update additionally require consequentialEffectAcknowledged; delete requires destructiveAcknowledged. All start unset. Expressions cannot grant consent. The original actor with current direct workspace AND organization membership must approve the exact account, method, body, retention and effects in a separate Human Task. Current account/secret, node/input, live policy, original decision, owned lease/checkpoint/native resume and one physical CAS claim bind dispatch and fresh OUTPUT. Used, expired or uncertain claims never resend.

Raw JSON, numeric lexemes, bounded records and unknown business fields enter normal history and downstream mappings. Links remain inert. Local caps remain30 seconds, 1MiB response, depth32,20,000 nodes,4,096 reads and6MiB output; malformed UTF-8/JSON, duplicate keys, invalid status/media or identity mismatch fail closed. Documented application/json; odata.metadata=minimal is accepted. Real Dataverse rights, provider effects and durable restart behavior remain unverified; local provider/socket/database tests are simulated.

Automatic import accepts only exact ordinary numeric version1 get/getAll/create/delete with explicit supported controls; Return All is refused. Original update requires manual migration because of its upsert semantics. Source ledgers retain parameters, versions and credential references without selecting an account or granting consent; export/reopen preserves invalid drafts until deliberate repair.

Primary contracts: retrieve a row, create a row, update and delete.

GA4 existing report token account

Choose GA4 and save a separately acquired workspace token in Named fields or Advanced JSON:

{ "profile": "ga4ReportManualBearerV1", "accessToken": "YOUR_ALREADY_ISSUED_ACCESS_TOKEN" }

Save, update and reopen stay encrypted, write-only and unverified with zero provider HTTP or DNS. Test connection, OAuth acquisition, refresh and Google Workspace token substitution are unavailable. Independently verify Data API enablement and property access, token scope/expiry and quota. App organization membership is distinct from GA4 property access; organization-scoped credential records are unsupported. Deleting, rotating or retyping the selected saved account cannot substitute another account by name. See the one-row raw report action. Live provider and Review behavior remain untested.

To replace a saved account, reopen it, re-enter the complete valid profile in Named fields or Advanced JSON and choose Update account. This updates the same account ID and keeps its replacement secret encrypted, write-only and unverified; reopening never reveals the saved token. Advanced JSON and Named fields share the exact profile validator.

The same GA4 account supports the broader bounded report.run profile without acquiring or refreshing a token. Check the metrics, filters, limits and source migration differences before selecting it.

Generic S3-compatible first page

Use the separate generic-s3 provider and genericS3ManualSigV4V1 account for a supplied public HTTPS root origin, region and explicit path-style boolean. This is distinct from the AWS S3 bucket provider. Named fields and Advanced JSON save the same encrypted, write-only account locally; Save/update/reopen perform no HTTP or DNS, never refill stored secrets and leave access unverified. Test, bulk Test, discovery, ambient credentials and refresh are unavailable. Manual-static omits token/expiry; manual-temporary requires a supplied token and exact UTC expiry with more than ninety seconds remaining at execution.

Only file.getAll.firstPage is available: select an explicit bucket, prefix, limit 1–1000 and saved account, then choose all three initially unchecked acknowledgments and the separate original-actor Human approval. One signed native TLS GET returns a bounded partial page without following continuation tokens. See the exact input, output and import limits. Real custom endpoint compatibility, account entitlement, native DNS/TLS, durable persistence and matching Review organization/workspace walkthroughs remain unverified.

Redis: external TLS string GET

The ordinary Redis subset reads one explicit string key with Redis → Get string key. It uses a separate external account; internal BullMQ/ioredis queue, lock and realtime credentials do not authorize it. Save redisManualTlsV1 with Named fields or Advanced JSON:

{
  "profile": "redis-manual-tls-v1",
  "host": "redis.example.com",
  "port": 6380,
  "database": 2,
  "username": "reader",
  "password": "YOUR_EXISTING_PASSWORD",
  "tls": true,
  "rejectUnauthorized": true
}

Enter the exact public DNS hostname, TLS port, numeric database and existing password. An explicit empty username selects password-only AUTH; a supplied username selects ACL AUTH. Save/update/reopen uses encrypted, write-only local storage and remains unverified, with zero DNS, HTTP or socket calls. Test and bulk Test are unavailable. No account, password or permission is acquired. Public IPv4 DNS answers, a pinned destination IP, original hostname/SNI and verified TLS are required; plaintext, verification disablement, private or IPv6-only routing, Unix sockets, Sentinel, cluster routing and discovery are unsupported.

Enable WORKFLOW_LIVE_REDIS=true and the current workspace Redis live preference independently. Select the exact saved account, set schema2/sourceVersion1 and deliberately review each initially unchecked migration, history-retention and one-attempt switch. The current original actor needs organization and workspace authority and separately confirms the resolved key, account revision and database in a signed Human Task. One current owned lease, stable decision, consumed original task/wait and physical CAS claim are rechecked through AUTH, SELECT, GET, response, settlement and fresh OUTPUT. A changed or consumed claim never authorizes another send.

The output preserves string text, including 001, whitespace and prototype-shaped business text. data.value:null with data.missing:true means a missing key; empty string means an existing empty value. data.output contains the literal selected output property. WRONGTYPE and authentication/protocol errors fail rather than returning null. Results may contain private data retained in ordinary workflow/checkpoint/Human history and downstream mappings or AI; account encryption adds no new raw-history encryption, retention deadline or scanner guarantee.

Local bounds are one GET after explicit AUTH/SELECT, thirty seconds total, one MiB of reply bytes, 128 reads, 64 KiB chunks and a 1,024-byte key. No PING, TYPE, INFO, KEYS, SCAN, enumeration, writes, retry, reconnect, offline queue, resend or returned-destination following occurs. A failed or unconfirmed close retains an uncertain outcome. Original version1 automatic type, dot notation, other operations, unknown options and Tool/Trigger intent remain unsupported and preserved for manual repair. Local tests use simulated Prisma/DNS/TLS facts; real Redis permissions/authentication, native network, durable DB/Redis restart and Review organization/Help walkthroughs remain untested. See GET and the RESP specification.

Xero: one existing-tenant contact read

The native ordinary numeric version1 subset is contact.get: one GET https://api.xero.com/api.xro/2.0/Contacts/{contactId} with an explicit UUID and Xero-tenant-id. Save the exact xeroContactManualTokenV1 JSON profile with tokenKind: "oauth2-access-token", manually supplied accessToken, existing tenantId and granted contactsScope (accounting.contacts.read or accounting.contacts). The external Xero tenant is distinct from the current application organization. Named fields and Advanced JSON Save/update/reopen are encrypted, write-only, local, unverified and send zero HTTP/DNS; Test and bulk Test skip. No OAuth refresh or tenant discovery.

All seven explicit boolean migration, output-retention, one-attempt, selected-account, tenant-authorization, approved-usecase and associated-user data-passthrough acknowledgments start unchecked and must be literal true. These user attestations never prove Xero approval or activate Live. The original actor separately decides the exact Human task; current organization AND workspace membership, account/material, node/input/policy, consumed decision, owned lease and one CAS physical claim are revalidated through persistence and fresh OUTPUT. Current provider approval, token/tenant entitlement and associated-user consent remain unverified; resolve those before actual provider activation/testing. Current Xero terms restrict approved use cases, third-party passthrough and model training/adaptation; they do not establish a blanket inference ban.

One exact contact in data and sourceItemsJson preserves bounded original business JSON, including normal prototype-shaped keys and possible personal, bank or tax information, in ordinary output/history/downstream. The raw envelope is discarded; returned URLs stay inert. Local30s/1MiB/fatalUTF8/structural bounds, zero redirects/retries and consumed receipts prevent automatic replay of uncertain attempts. Import/export/reopen retain unsupported raw intent, raw versions and credential IDs; no automatic consent or OAuth-credential alias. Invoice, list, writes, Tool/Trigger and fullnode parity remain unsupported. Real provider, DNS/TLS, Review, durable DB/restart and real human authentication remain untested.

Consent must be saved as seven literal true booleans; acknowledgment expressions are unavailable and cannot acquire consent by resolving to true. A timely settled contact can pass through a later Wait under a new trusted resume revision and current owned lease, with its original signed task/wait/decision/claim/result conserved and current authority checked again; OUTPUT never sends again. The original send approval expires after15minutes; a timely signed settled result may outlive that send window, while retries, recovery or changed checkpoint attempts refuse.

Azure Storage: first container page

Use Azure Storage → List containers (first page), operation container.getAll / tool azureStorage_container_getAll. This is an ordinary numeric n8n typeVersion 1 compatibility subset. Save an existing account in Named fields or Advanced JSON:

{
  "profile": "azureStorageManualBearerV1",
  "account": "youraccount",
  "accessToken": "YOUR_EXISTING_STORAGE_BEARER",
  "audience": "https://storage.azure.com/",
  "expiresAt": "2099-01-01T00:00:00.000Z"
}

Use the token's actual expiry; the example date is a placeholder. Account is 3–24 lowercase letters/digits, without a URL. Save/update/reopen encrypt locally, are write-only and remain unverified with zero provider HTTP. There is no Test, OAuth acquisition/refresh, SharedKey signer, SAS, discovery or Azure resource creation. Token signature/issuer/audience/expiry claims and effective rights remain unverified. List Containers permissions require Microsoft.Storage/storageAccounts/blobServices/containers/read scoped to the account or above; a blob-level entitlement alone does not prove account-list access. Tokens expiring within 30 seconds are refused at execution.

Original imports must declare exactly ordinary numeric version 1, resource:container, operation:getAll, returnAll:false, limit:1, and matching original authentication:oAuth2/azureStorageOAuth2Api or authentication:sharedKey/azureStorageSharedKeyApi. Original declarations and unsupported intent remain in the source ledger. Reselect the separate manual bearer account; no original OAuth/SharedKey authentication parity is claimed. Other defaults/options, metadata/system/deleted filters and Tool/Trigger require visible manual repair; additional admitted operations are described below.

Input requires parameterSchemaVersion:2 and four independent literal acknowledgments: migrationAcknowledged:true, rawRetentionAcknowledged:true, onePhysicalAttemptAcknowledged:true and firstPageOnlyAcknowledged:true; none is automatically checked. The original actor needs current direct organization AND workspace membership, current selected account and input revisions, WORKFLOW_LIVE_AZURE_STORAGE=true and the independent current workspace live preference. A separate signed Human Task binds the exact safe request preview without embedding the bearer. Nonempty owned lease, original task/wait/decision, checkpoint attempt and consumed resume revision remain required. One CAS claim permits one physical attempt; uncertain or used claims never resend, including fresh resumes or node retry settings.

The request is exactly GET https://{account}.blob.core.windows.net/?comp=list&maxresults=1, with Bearer authorization, a fresh UTC x-ms-date, and source x-ms-version:2021-12-02. Public DNS, admitted IP, certificate/SNI and remote peer are revalidated. For this first-container-page action, custom/private/sovereign endpoints, alternate ports, redirects, retries and continuation requests are unsupported; IPv6-only hosts remain unsupported.

A successful 200 XML response yields data.containers, data.maxResults:1 and opaque data.nextMarker. paging.completeness is partial-first-page when a marker exists, otherwise complete-first-page; this never implies full pagination. Empty pages retain their marker, fixing the original helper shortcut's loss. Unknown container fields/attributes and textual metadata (including 0007) persist in normal history and downstream mappings. Returned URLs remain inert. Root-level unknown fields are omitted as in the original helper. XML names follow source helper normalization; strict named boolean/safe integer conversion applies outside metadata. Prototype-sensitive names and normalization collisions refuse success. This differs deliberately from the original metadata coercion.

Limits: one total 30-second dispatch/read/decode deadline, 256 KiB body, depth 24, 4,096 elements/reads, one container, and an 8,192-character marker. Strict UTF-8/XML, exact response status/media/length, secret-reflection checks and bounded streaming precede success. DTD, custom/external entities, namespaces, comments, CDATA and processing instructions are rejected. The five predefined XML entities and valid numeric character references are decoded without external resolution. Raw transport text is discarded; no full XML/parser parity is claimed. Existing business data may be private and remains in ordinary history; this adds no retention/encryption guarantee for that history.

Local tests simulate provider responses, Prisma facts and native sockets. Real Azure token claims/entitlement/RBAC propagation, service responses, network ACL/private-network constraints, native deployment, actual durable database/Redis restart and organization-scoped Review/tutorial behavior remain untested.

Azure Storage and Cosmos DB: additional bounded operations

Azure Storage also exposes container.get/create/delete and blob.getAll/get/create/delete, using the same exact azureStorageManualBearerV1 account and fixed public origin. Select the operation, review its disclosure and supply schema 2 plus literal migration, retention and one-attempt consent. Container creation needs createContainerAcknowledged; container deletion needs deleteContainerAcknowledged and removes every blob. Blob upload needs replaceBlobAcknowledged, explicit UTF-8 content (at most 4 KiB) and contentType; it can replace content and metadata. Blob deletion needs deleteBlobAcknowledged; snapshots, leases and conditional requests are unsupported. Every write requires a separate exact-request Human Task and one physical claim; a lost response never authorizes retry.

blob.getAll requires firstPageOnlyAcknowledged and an explicit limit of 1–100. Optional include accepts only the fields exposed by the form. It returns data.blobs, data.maxResults and an opaque data.nextMarker; empty partial pages retain their marker and it is never followed. blob.get requires textOnlyAcknowledged and returns bounded UTF-8 data.content with name, properties and metadata. Original n8n blob downloads produce binary items and remain refused; configure this different text-only operation manually. Original blob uploads using binary or URL sources also remain refused. Container read/create/delete, bounded blob listing and deletion can migrate only their admitted literal source options. Unsupported fields remain visible for repair. XML keeps the 256 KiB/depth-24/4,096-element bounds; text download is at most 256 KiB. All calls retain the total 30-second budget.

Azure Cosmos DB also exposes container.getAll/create/delete and item.get/getAll/query/create/update/delete with the exact azureCosmosDbMasterKeyV1 account, saved database and fixed public origin. Every operation requires schema 2 and literal migration, retention and one-attempt consent; all five writes additionally require writeEffectAcknowledged. The original actor needs current organization AND workspace membership and both live gates, then confirms the exact selected account, request and body. Every action uses one HMAC-authenticated request and one physical claim; uncertainty remains unverified and never resends.

Container/item listing accepts one page with maxItemCount 1–100; continuation is opaque and never followed. Query sends one POST with explicit SQL and named parameters, not a write. Choose crossPartition (potentially higher RU charges) or singlePartition with an explicit string partition key. It accepts at most 1,000 returned values; a page may be empty or partial. Point item reads and writes need an explicit string partition key; no discovery probe or inferred key is used. Return values are data.containers, data.items, data.item, data.container or {deleted:true}, with bounded envelopes and numeric-lexeme metadata. New item/query responses preserve otherwise unsafe numeric spellings as text. The original container.get contract retains its stricter numeric refusal. Normal history and downstream mappings retain business data.

Native item creation never upserts; an existing id conflicts. Native item update replaces the whole item using the exact authored JSON and matching id; omitted properties disappear. Both differ from original n8n upsert/injected-body behavior, so those imports remain refused and require manual configuration. Container creation accepts one Hash partition path and optional indexing policy/manual throughput: 400–1,000,000 RU/s in steps of 100 can incur ongoing billing. Deleting a container removes all contents. No patch/upsert/autoscale, retries, redirects or automatic continuation is implemented. Limits remain total 30 seconds, 1 MiB response, depth 32 and 20,000 nodes; authored item/policy JSON is at most 8 KiB.

These are bounded native operations and explicit migrations, not full n8n item/authentication parity. Local HOST tests use simulated provider responses and database facts. Real provider credentials, grants, charges, native DNS/TLS, durable restart and authenticated organization Review/tutorial behavior remain untested.

Firebase Realtime Database supplied-token account

Choose Firebase Realtime Database and create an exact firebaseRealtimeManualBearerV1 workspace account. Named fields and Advanced JSON use the same parser. Named fields record the two required scope metadata values; verify those scopes and the actual token kind/permissions separately. The database instance name is not necessarily the Google project ID. Copy the exact origin from your database configuration; do not infer the suffix.

{
  "profile": "firebaseRealtimeManualBearerV1",
  "databaseOrigin": "https://YOUR_DATABASE_INSTANCE.firebaseio.com",
  "accessToken": "YOUR_ALREADY_ISSUED_GOOGLE_OAUTH_ACCESS_TOKEN",
  "expiresAt": "2099-01-01T00:00:00.000Z",
  "scopes": ["https://www.googleapis.com/auth/userinfo.email", "https://www.googleapis.com/auth/firebase.database"]
}

Replace the example expiry with the supplied token's actual UTC expiry. Supported finite suffixes are firebaseio.com, europe-west1.firebasedatabase.app and asia-southeast1.firebasedatabase.app; no custom/private origins or emulators. Save/update/reopen use existing encrypted write-only storage, remain unverified, and make zero HTTP/DNS calls. No Test connection, OAuth acquisition/refresh, Firebase ID-token auth query, legacy secret, anonymous or generic Google Workspace credential reuse. The token may have broad rights: Google service-account OAuth tokens can bypass Database Rules. App organization membership does not establish Firebase permissions. Account records scoped to organizations are unsupported; the original actor must still belong to both app organization and workspace. Actual token grants, database existence, network access and quota/bandwidth/charges remain unverified. See the one-path read.

Explicit ordinary numeric version 1 Firebase update, push, create and delete now migrate to separate bounded native writes: PATCH, POST, PUT and DELETE respectively, using the same firebaseRealtimeManualBearerV1 account and one nonroot literal path. The omitted original default is still refused. Body operations retain the original comma-separated attributes; bind inputData explicitly to the upstream record because n8n implicitly read the current item. Only listed own top-level properties present in that record are sent, as canonical JSON (at most 4 KiB). Requests with no matching attributes, server values or multi-path keys are refused.

All writes require the three common literal acknowledgments plus externalWriteAcknowledged; PUT also needs fullReplacementAcknowledged because omitted subtree values are removed, and DELETE needs subtreeDeletionAcknowledged because it removes the entire subtree. Import leaves every consent false and does not select or acquire credentials. Review the exact account, verb, path and body in the original-actor Human Task under current organization AND workspace authority and both live gates. One durable physical claim permits one attempt; a timeout, disconnect or malformed response leaves the outcome unverified with no automatic retry or replay. Successful writes retain bounded data, rawJson and request-body byte count in ordinary history. POST returns the generated child name; DELETE returns null. These are explicit adaptations; original scalar/null/item wrapping and implicit per-item execution parity are not claimed. Real provider writes, charges and Review walkthroughs remain untested.

QuickBooks: one bill with an existing manual token

Only exact ordinary n8n-nodes-base.quickbooks, numeric typeVersion: 1, explicit resource: "bill" and operation: "get" are supported. Save the separate quickbooksBillManualBearerV1 account using Named fields or Advanced JSON:

{
  "profile": "quickbooksBillManualBearerV1",
  "tokenKind": "oauth2-access-token",
  "accessToken": "YOUR_EXISTING_ACCESS_TOKEN",
  "realmId": "1231434565226279",
  "environment": "production",
  "accountingScope": "com.intuit.quickbooks.accounting",
  "expiresAtEpochMs": 1893456000000
}

Replace the example realm and expiry with the already authorized company and actual token expiry metadata. Environment must be explicitly production or sandbox; scope must be exactly com.intuit.quickbooks.accounting, and expiry is a positive safe integer in epoch milliseconds. Save is encrypted, write-only, local and unverified, with no provider HTTP/DNS or Test. Reopen an account, re-enter the complete replacement profile in Named fields or Advanced JSON and choose Update account to preserve its ID; saved token text is never shown. There is no OAuth acquisition, refresh, implicit conversion from quickBooksOAuth2Api, discovery or generic Google credential reuse. Scope, expiry and realm declarations do not verify grants, company access, app approval or quota. The external company realm is distinct from the application's organization and workspace.

Enter one exact string billId. Realm and bill IDs use a local subset of 1–40 ASCII decimal characters, retaining leading zeros and IDs above JavaScript's safe integer range as text; this is not the complete Intuit ID grammar. The initially unchecked acknowledgeFinancialDataPersistence must be saved as literal true; expressions and truthy strings fail before rendering and on restored WAIT. A separate original-actor Human task confirms the account and exact read. Current organization AND workspace authority, selected account revision/material, realm/environment, decision, lease, checkpoint and one physical CAS claim bind dispatch and persistence. A saved account or acknowledgment alone grants no execution authority.

The request is one GET /v3/company/{realmId}/bill/{billId}?minorversion=75 on fixed https://quickbooks.api.intuit.com or https://sandbox-quickbooks.api.intuit.com, with no body, redirects or retries. Version75 is the documented floor/default, not a frozen complete schema. Local refusal bounds include one thirty-second total deadline, one MiB response, fatal UTF-8, depth32 and20,000 JSON nodes. Expired metadata, changed authority/account, cancellation, unsafe JSON or uncertain used claims refuse success; fresh OUTPUT validates the settled receipt and current authority without sending again.

An open Bill object must contain the exact requested string Id; TotalAmt and SyncToken are not required by this subset. Unknown nested fields and false/null values remain business data. Parsed Bill data and exact rawResponseJson, including financial/contact data, enter ordinary workflow result/history/downstream persistence. Raw JSON retains decimal number spelling; parsed JavaScript numbers do not promise exact financial arithmetic. Duplicate keys, unsafe integer tokens, missing/null/array/wrong-ID Bill, any Fault including HTTP200, and explicit credential-value echoes fail. This adds no history retention deadline, encryption or scanner guarantee.

Import/export/reopen/repair preserve original versions, credential references, unsupported options and malformed, duplicate or whole-input intent for explicit manual repair. Other resources/actions, list/query/paging, writes, Tool/Trigger/fuzzy aliases, custom origins/minor versions and complete node/Bill-schema parity are unsupported. Local tests use synthetic credentials and simulated provider, storage and socket facts. Real Intuit access, native DNS/TLS, durable database/restart and authenticated Review organization/Help walkthroughs remain untested. See Intuit minor versions and OAuth authorization.

CrateDB and TimescaleDB manual TLS read

The separate native-pg-read provider uses the exact nativePgReadManualTlsV1 saved-account type. Choose profile:"crateDb" or profile:"timescaleDb" explicitly and supply host, numeric port, database, user, write-only password, tlsProfile:"system", serverName equal to host, and the exact reviewedSchema, reviewedTable and reviewedColumns string array. The host must be a public DNS hostname. No connection string, custom CA or key path, insecure TLS, SSH, discovery or credential refresh is supported. Database and user are ASCII non-space strings of 1–63 characters; the password is ASCII, up to 1024 characters. Authentication supports trust, TLS cleartext password and SCRAM-SHA-256; MD5 is unsupported. Account creation does not verify database permissions, server capabilities or relation semantics.

Named fields and Advanced: edit JSON produce the same account JSON. For example:

{
  "profile": "crateDb",
  "host": "pg.example.net",
  "port": 5432,
  "database": "doc",
  "user": "reviewed_reader",
  "password": "supplied-write-only-password",
  "tlsProfile": "system",
  "serverName": "pg.example.net",
  "reviewedSchema": "doc",
  "reviewedTable": "contacts",
  "reviewedColumns": ["id", "name"]
}

Save, update and reopen are local, encrypted, write-only and unverified. They perform no DNS lookup, provider HTTP or database connection. Test and bulk Test are unavailable. Reopen displays account identity and leaves stored material empty; enter the complete replacement to update the same saved account ID. Replacing material invalidates earlier approval. Original n8n credential references remain source provenance and never choose a saved account.

Only query.execute / native_pg_read_query_execute is available. See the literal SQL, input and output contract and exact import repair rules. Current workspace and organization access and a separate original-actor Human approval are required. Actual account/provider/schema/role compatibility, native deployment DNS/TLS, durable restart and the real Settings/Review walkthrough remain untested until legitimate credentials and the environment are available.

Cuenta manual y alcance en español

Selecciona nativePgReadManualTlsV1, el perfil exacto crateDb o timescaleDb, el host DNS público, puerto numérico, base de datos, usuario y contraseña de solo escritura. Usa tlsProfile:"system", serverName igual al host y el esquema, tabla y columnas revisados. Los campos y el editor JSON guardan la misma cuenta cifrada; actualizar conserva su ID e invalida aprobaciones anteriores. Guardar y reabrir no conectan ni verifican permisos; no hay Test, DNS, descubrimiento ni renovación. Al reabrir, el material guardado permanece vacío. Las referencias originales no eligen una cuenta. La compatibilidad real, DNS/TLS nativos, reinicio duradero y el recorrido de Settings/Review siguen sin probarse.

A blank Named password explicitly saves password:"", including on update; it never retains the previous password. Advanced JSON can supply the same empty string. En español: dejar vacía la contraseña guarda password:"" explícitamente, también al actualizar; nunca conserva la contraseña anterior.

LDAP manual LDAPS base-search account

The preview native-ldap-read integration uses the separate nativeLdapReadManualLdapsV1 account. Supply this complete JSON through Named fields or Advanced JSON:

{
  "security": "ldaps",
  "trust": "system",
  "host": "ldap.example.com",
  "port": 636,
  "bindDN": "cn=reader",
  "password": "REPLACE_WITH_EXISTING_PASSWORD",
  "reviewedSelection": {
    "baseDN": "cn=target",
    "scope": "base",
    "returnAll": false,
    "limit": 1,
    "paged": false,
    "attributes": ["cn", "mail"],
    "filter": {
      "kind": "equality",
      "attribute": "cn",
      "value": "target"
    }
  }
}

Use a public DNS hostname, fixed port636, system trust and an exact literal bind DN. The saved reviewed selection must match the one invoked base DN, projected text attributes and equality AST. Passwords are write-only. Save and update validate locally, encrypt the complete account, return masked summaries and make no directory/DNS request. Reopen does not reveal saved material; re-enter a complete replacement. Replacement preserves the ID and invalidates earlier approvals. Direct and bulk Test are unavailable. Directory access, grants and server semantics remain unverified.

Plain LDAP, StartTLS, custom CA/key files, SASL/anonymous bind, discovery, paging, subtree search, binary attributes, DN normalization/escaping and imported account authority are unsupported. This is an explicit ldapts9.2.0 migration from original4.2.6 with no complete LDAP-node parity claim. See bounded input/output and original import.

En español: usa la cuenta independiente nativeLdapReadManualLdapsV1, host DNS público, puerto636, LDAPS con confianza del sistema, DN de enlace y selección base revisada exactos. Guardar y actualizar validan y cifran localmente sin conectar; reabrir no muestra secretos y sustituir la cuenta completa mantiene el ID e invalida aprobaciones. No hay Test ni descubrimiento. Los permisos y el directorio real siguen sin verificar.

MySQL manual TLS reviewed row

Choose MySQL reviewed row and save the exact nativeMySqlReadManualTlsV1 account with serverProfile:"mysql84", a public DNS host, numeric port, literal database, user, write-only password, tlsProfile:"system", serverName equal to the host, reviewedTable and ordered reviewedColumns JSON array. Use a database account limited to SELECT on that table. Save validates and encrypts locally without checking privileges, connecting, resolving DNS or discovering schemas. Test and bulk Test are unavailable. A complete replacement keeps the account ID and invalidates previous approvals; reopened fields never reveal saved secrets. No connection URL, SSH, custom trust, plaintext, pool or reconnect is provided. This bounded profile permits supported single-factor authentication; an actual additional-factor exchange is refused even when a server advertises MFA capability.

Select the current account explicitly after import, set the three literal migration/raw-retention/one-attempt acknowledgments and complete the separate exact Human approval as the original actor with current organization and workspace membership. One confirmed physical attempt has no retry or fallback. See the finite input and output contract. Your server, grants, public DNS and authenticated Review walkthrough remain unverified by an owned local protocol test.

En español: elige nativeMySqlReadManualTlsV1, mysql84, host DNS público, puerto, base de datos, usuario, contraseña de solo escritura, TLS del sistema con nombre igual al host y tabla/columnas revisadas. Guardar o reemplazar valida y cifra localmente, sin conectar ni probar permisos; reabrir no revela secretos y no hay Test. Selecciona la cuenta y las tres confirmaciones literales; otra aprobación Human del actor original exige acceso vigente a la organización y al workspace. Solo se permite un intento, sin reintentos.

MongoDB manual system-TLS document-read account

The preview native-mongo-read integration uses nativeMongoReadManualTlsV1. Supply the complete JSON through Named fields or Advanced JSON:

{
  "profile": "mongoDb",
  "host": "mongo.example.com",
  "port": 27017,
  "database": "app",
  "authSource": "app",
  "user": "reader",
  "password": "REPLACE_WITH_EXISTING_ASCII_PASSWORD",
  "authMechanism": "SCRAM-SHA-256",
  "tlsProfile": "system",
  "serverName": "mongo.example.com",
  "declaredReadOnly": true,
  "reviewedCollection": "records",
  "reviewedFields": ["label", "amount"]
}

Use a public lowercase DNS hostname and exact matching serverName/authSource. No URL/SRV, custom trust, proxy, referral/discovery, implicit credential acquisition or live Test. Save/update are local, encrypted, write-only and unverified; enter a complete replacement. The read-only declaration does not verify grants. Import preserves source intent without selecting a saved account. Select three literal acknowledgments and complete the separate exact original-actor Human decision before one find. Fresh organization AND workspace membership, account/source/policy revision and durable lease are required; used or uncertain claims never replay. Canonical EJSON output and raw business data persist in authorized history and may reach downstream AI. See the MongoDB in-app guide for field/output limits. Real provider credentials, deployment and Review tutorial checks are still required.

En español: introduce el perfil completo de cuenta manual y la colección/campos revisados. Guardar y actualizar no conectan ni verifican permisos; reabrir no revela secretos. No se admiten URL/SRV ni TLS inseguro. Importar no selecciona cuenta ni confirmaciones. Una decisión Human firmada exige acceso vigente a organización Y espacio. Los datos EJSON se conservan en el historial autorizado; los intentos usados o inciertos no se repiten. Las pruebas de proveedor y Review siguen pendientes.

Jina Reader manual key

The preview jina-reader provider uses only jinaReaderManualKeyV1. Enter complete JSON through Named fields or Advanced JSON:

{
  "apiKey": "REPLACE_WITH_EXISTING_MANUAL_JINA_KEY"
}

Save/update validate locally, encrypt a complete replacement and return masked summaries. Reopen is write-only; replacement retains the ID and invalidates prior approval. The key profile accepts 8–4096 ASCII bearer-token characters (letters, digits, dot, underscore, tilde and hyphen); other key formats refuse this finite native profile. Direct and bulk Test are unavailable: no provider, balance, dashboard query or grant acquisition occurs. Imported jinaAiApi references stay inert. See bounded Reader block and original import.

Jina performs the remote target fetch; its target DNS, redirect and access-policy behavior are not local transport facts. One request can disclose the target URL and consume credits. Raw untrusted page data persists in authorized workflow history and may reach downstream AI. Provider/key compatibility, credits, deployment and authenticated Review remain untested. Fresh authority checks and guarded writes do not make database and network actions atomic.

En español: Jina obtiene el destino remoto; su DNS, redirecciones y políticas no son hechos del transporte local. Una solicitud puede revelar la URL y consumir créditos. Los datos sin confianza persisten en historial y pueden llegar a IA posterior. Compatibilidad real, créditos, despliegue y Review autenticado siguen sin probarse. Organización Y espacio, actor original y cuenta guardada vigente siguen siendo obligatorios; un intento usado o incierto no se repite.

AWS Rekognition DetectLabels (preview)

Save an existing manual AWS account using awsRekognitionManualSigV4V1 and profile aws-rekognition-manual-sigv4-v1 in us-east-1, then explicitly select it on the imported DetectLabels block. Account saving is local and unverified; Test and bulk Test send nothing to AWS. Select the exact S3 object version and review all six initially unchecked acknowledgments before requesting approval. AWS image retention and local run history are separate: full bounded raw JSON can reach downstream AI.

Execution requires both the deployment setting WORKFLOW_LIVE_NATIVE_REKOGNITION=true and the workspace live preference for aws-rekognition. Human approval belongs to the original actor and binds the saved account, exact source, input and workflow version. Changes require a new approval. An uncertain paid outcome must not be replayed.

Real AWS requests and authenticated organization/workspace Review walkthroughs remain untested. Local simulated responses do not verify AWS permissions, image access or provider retention settings. See DetectLabels fields, acknowledgments and limits.

Español: DetectLabels en vista previa

Guarda una cuenta AWS manual existente con awsRekognitionManualSigV4V1 y perfil aws-rekognition-manual-sigv4-v1 en us-east-1, y selecciónala explícitamente en el bloque DetectLabels importado. Guardar es local y no verifica la cuenta; las pruebas individuales y masivas no envían nada a AWS. Indica la versión S3 exacta y revisa las seis confirmaciones inicialmente desmarcadas antes de solicitar aprobación. La retención de imágenes en AWS y el historial local son independientes: el JSON bruto completo y acotado puede llegar a IA posterior.

La ejecución exige tanto WORKFLOW_LIVE_NATIVE_REKOGNITION=true en el despliegue como la preferencia de ejecución real de aws-rekognition en el workspace. La aprobación humana corresponde al actor original y vincula la cuenta guardada, el origen exacto, la entrada y la versión del flujo. Los cambios exigen una nueva aprobación. Un resultado facturable incierto no debe repetirse.

Siguen sin probarse las solicitudes reales a AWS y los recorridos autenticados de organización y workspace en Review. Las respuestas locales simuladas no verifican permisos de AWS, acceso a la imagen ni la política de retención del proveedor. Consulta los campos, confirmaciones y límites de DetectLabels.

AWS Textract AnalyzeExpense (preview)

Create the manual Textract AWS account, save it and explicitly select it on the AnalyzeExpense block. Save keeps keys encrypted and write-only, does not contact AWS and leaves the account unverified. Test and bulk Test are unavailable. Reopening leaves keys blank; updating requires complete replacement. Temporary accounts need a supplied token and UTC expiry with more than ninety seconds remaining at dispatch.

Use one original input item containing one ready workspace PNG or baseline JPEG, no larger than 256 KiB or 4096 pixels on either side. This checks a finite image structure; it does not decode pixels or establish what AWS will recognize. Review the five separate migration, possible-charge, provider-retention, local raw-retention and full-image-disclosure acknowledgments. All start unchecked. They do not replace the original actor's exact Human approval.

The entire base64 image request and complete expense response can enter normal approval/execution history and downstream AI. No additional history encryption, expiry or retention opt-out is promised. Provider retention settings are separate and are neither verified nor changed by consent.

Execution requires both the deployment setting WORKFLOW_LIVE_NATIVE_TEXTRACT=true and the workspace live preference aws-textract. Real AWS requests and authenticated organization/workspace Review walkthroughs remain untested.

Español: AnalyzeExpense en vista previa

Crea la cuenta AWS manual de Textract, guárdala y selecciónala explícitamente en el bloque AnalyzeExpense. Guardar conserva las claves cifradas y sin volver a mostrarlas, no contacta AWS y no verifica la cuenta. Test y Test masivo no están disponibles. Al reabrir quedan vacías; actualizar exige reemplazarlas por completo. Las cuentas temporales necesitan token aportado y caducidad UTC con más de noventa segundos de vigencia al ejecutar.

Usa un único elemento original con un PNG o JPEG baseline listo del workspace, de hasta 256 KiB y 4096 píxeles por lado. Se comprueba una estructura limitada, sin decodificar píxeles ni garantizar lo que reconocerá AWS. Revisa las cinco confirmaciones separadas de migración, posible cobro, retención del proveedor, retención local sin procesar y divulgación de la imagen completa. Empiezan desmarcadas y no sustituyen la aprobación Human exacta del actor original.

La solicitud con la imagen completa en base64 y la respuesta completa pueden entrar en historial ordinario y llegar a IA posterior. No se promete cifrado adicional, caducidad ni exclusión de retención del historial. El consentimiento no verifica ni modifica la retención configurada en AWS.

Se exige tanto WORKFLOW_LIVE_NATIVE_TEXTRACT=true en el despliegue como la preferencia real aws-textract del workspace. Siguen sin probarse las solicitudes reales AWS y los recorridos autenticados de organización y workspace en Review.

Microsoft SQL manual account (storage only)

Choose Microsoft SQL in Settings → Integrations to store a separate manual account with credential type microsoftSqlManualTds74TlsV1 and provider microsoft-sql. Enter exactly profileId: "mssql-manual-tds74-tls-v1", host, numeric port, database, username and password, using named fields or Advanced JSON. There are no inferred host or port defaults. Save validates locally, encrypts the complete profile and keeps it write-only and unverified. Reopen leaves the fields blank; Update account requires all six fields and replaces the complete profile.

Use an explicit DNS or IPv4 host, an integer port from 1 to 65535, and SQL username/password authentication. Named instances, connection URLs and Windows domain usernames are unsupported. Password spaces, tabs and Unicode remain exact; passwords containing CR or LF stay in Advanced JSON with escaped line breaks. Conversion to named fields refuses values that would change. The secret contains no account ID, ciphertext revision, reviewed selection or client authority flags.

No SQL operation or SELECT execution is available. Test and bulk Test are unavailable; Save makes no DNS, connection, SQL or schema-discovery request. The account remains unverified: server identity, database permissions, grants and row-level security have not been checked. Saving it does not repair an imported node; see Microsoft SQL import limits.

Español: cuenta manual de Microsoft SQL (solo almacenamiento)

Elige Microsoft SQL en Configuración → Integraciones para guardar una cuenta manual independiente con tipo microsoftSqlManualTds74TlsV1 y proveedor microsoft-sql. Introduce exactamente profileId: "mssql-manual-tds74-tls-v1", host, port numérico, database, username y password, en los campos o en JSON avanzado. No se infieren valores predeterminados de host ni puerto. Guardar valida localmente, cifra el perfil completo y no vuelve a mostrarlo ni lo verifica. Al reabrir, los campos quedan vacíos; Actualizar cuenta exige los seis campos y reemplaza el perfil completo.

Usa un host DNS o IPv4 explícito, un puerto entero de 1 a 65535 y autenticación SQL con usuario y contraseña. No se admiten instancias con nombre, URL de conexión ni usuarios de dominio de Windows. Los espacios, las tabulaciones y Unicode de la contraseña permanecen exactos; las contraseñas con CR o LF quedan en JSON avanzado con los saltos de línea escapados. La conversión a campos rechaza valores que cambiarían. El secreto no contiene ID de cuenta, revisión del texto cifrado, selección revisada ni indicadores de autoridad del cliente.

No hay operaciones SQL ni ejecución de SELECT disponibles. Test y Test masivo no están disponibles; guardar no hace solicitudes de DNS, conexión, SQL ni descubrimiento de esquemas. La cuenta permanece sin verificar: no se han comprobado la identidad del servidor, los permisos de base de datos, las concesiones ni la seguridad por filas. Guardarla no repara un nodo importado; consulta los límites de importación de Microsoft SQL.

OpenAI accounts in workflow blocks

For a deployed destination, its saved connection bindings take precedence over the node's selected account. A notice beside the selector names a different deployed account and offers Edit destination connections. Step results record the resolved credential name and its source, including on provider errors. Changing the node selector alone does not change the deployed binding. See destination connections before retrying a live step with an account that has different permissions or credits.

In an Agent or Embeddings node, open Credential → + Add account.... In a generic HTTP Request, choose Authentication → Predefined Credential Type, select OpenAI as Credential Type, then add or select an Account. The form stores an API Key, optional Organization ID and Base URL as an encrypted workspace credential. Select the same account in other OpenAI blocks to reuse it. Account secrets stay out of the workflow graph; imported n8n accounts require their key to be configured in Confluye before a live request can run.

Google language full-response compatibility

The additive operations google-translate / translate.n8nV1 and google-natural-language / analyzeSentiment.n8nInlineV1 use the existing matching manually saved bearer accounts and workspace live controls. Save remains local and unverified. No OAuth login, token acquisition/refresh, source service-account execution or provider Test is added. Select the current account explicitly; imported credential references remain inactive. Workspace and organization authorization and separate gateway confirmation still apply.

Both require exact text, actual boolean acknowledgeTextDisclosure, acknowledgePossibleBilling and acknowledgeFullResponseRetention, all starting unchecked. Translate also requires targetLanguage. The last choice permits retaining the whole provider response object, including nested personal text and unknown fields, in normal workflow configuration, execution history and downstream AI data. It does not establish Google's retention policy or account consent. The gateway independently confirms the exact request. Failure after dispatch may already have incurred processing and charges; no automatic retry follows.

Translate compatibility omits the format request property and preserves the complete first translation, without decoding HTML/entities. The existing plain-text translate operation is unchanged. Natural Language compatibility sends inline PLAIN_TEXT, UTF16 and autodetection, preserving sentences and numeric metrics/offsets. The existing analyzeSentiment selected-output operation still returns metric strings. No medical or safety interpretation is added.

The full-output contract accepts finite JSON within 256 KiB, depth 32 and 32768 nodes, with string/key budgets, after a 1 MiB bounded response read. It refuses credential-shaped keys, unsafe integers, decimal precision loss, negative zero, numeric overflow/underflow and malformed output as a whole; it never silently drops unknown business fields. Returned URLs are not followed. The same 5000-code-point/10000-unit/32768-request-byte text limits, fixed HTTPS origins, deadlines and one-attempt policy apply. The import profiles describe exact versions, unsupported controls and one-input-item admission. Real Google execution and authenticated Review workspace/organization flows remain untested; local transport and consent responses are simulated.

Google directory bounded raw getters

The additive Google Chat spaces.getRaw and Google Contacts contacts.getRaw operations preserve a complete bounded response. The existing selected-metadata getters and page operations keep their previous schemas. Use the same exact googleChatManualBearerApi or googleContactsManualBearerApi saved account, containing only accessToken, and the existing matching workspace live control. Save is local and unverified; Test, OAuth acquisition/refresh and service-account conversion remain unavailable. Imported credential references never select or authorize an account.

Chat takes a literal resourceName matching spaces/[A-Za-z0-9_-]{1,256} and sends one GET https://chat.googleapis.com/v1/{resourceName} without a field mask. Contacts takes a literal people/{suffix}, excluding people/me, and a unique, nonempty personFields subset of names, emailAddresses, phoneNumbers, organizations, birthdays. It sends one GET https://people.googleapis.com/v1/people/{suffix}?personFields=... in the selected field order. Both correlate the returned identity. Contacts preserves etag, metadata and nested fields and adds contactId from the verified resource name.

Deliberately select the initially unchecked acknowledgeFullRecordDisclosure control, then review independent execution confirmation. Complete private records enter ordinary retained run/checkpoint history and downstream items. Audit telemetry contains operation and attempt metadata, without record contents. No automatic TTL or additional database encryption is promised. The same current account, actor, workspace/organization authority and live checks apply as for the existing manual readers.

Exactly one actual input item is required before dispatch. Each operation makes one request attempt, including when block retries are enabled, without redirects, automatic pages, discovery, warmups or returned-link fetching. Local limits are 1 MiB decoded response, 256 KiB serialized output, 131072 aggregate string units, depth 24, 8192 value/key visits, 256 units per key and 65536 per string. JSON numbers must serialize without a change in numeric value; precision loss, unsafe integers, nonfinite values, nonzero underflow and negative zero refuse the entire record. Duplicate/prototype-shaped keys, embedded secret fields, reflected current credentials, invalid Unicode, wrong identities and overflows fail completely, without dropping fields or truncating content. Provider and gateway budgets remain 10 and 11 seconds after preparation; preparation has no bounded deadline.

Automatic migration supports only ordinary numeric n8n version 1: Chat explicit authentication:"oAuth2", resource:"space", operation:"get", literal spaceId; or Contacts resource:"contact", operation:"get", literal contact suffix, explicit rawData:true and the supported unique field subset. Migration preserves the original disabled state and immutable source ledger, clears imported account authority and starts retention consent false. Lists, clean/default Contacts mode, wildcard fields, extra options, expressions, retries/continue-on-error, Tool/Trigger/HitlTool, messages, members, media, webhooks, wait and writes stay refused. These are two bounded executable profiles; real Google permissions, provider responses, human consent and authenticated Review organization/workspace behavior remain unverified.