Workspace settings

Workspace Settings lives at /{slug}--{workspaceId}/settings/{section}. Default section is General. Deep links accept a resource id after the section (for example an API key id).

Navigation

Authentication contains enterprise organization and SSO settings. Environments preserves previous environment and release records, with links to each workflow's Versions & previews; it does not publish new releases (Enterprise). Environments also shows Private credentials (1Password): a workspace Owner connects one vault and grants a credential to a workflow and an end user (only runs for an end user can use it, so it needs end users enabled), and other roles see only a note that it needs an Owner (Credentials from a private 1Password vault). Mailer is documented under Mailer. API keys: API Keys. Security is where you verify MFA before AIMS governance actions. Approvals, in the main navigation, is the queue-first AIMS surface for governed drafts, review, renewal, and blocked production — see AI governance.

The previous AI Governance (governance) URL remains valid and opens the same Approvals page outside the Settings layout. Its Pending, Workflows, Approvals & history, and Configuration tabs preserve their links. The entry appears for an effective interactive AIMS assignment, an eligible setup step, or an Organization Owner who can manage approval policy; workspace ownership alone does not grant AIMS access.

General: workspace identity

The workspace directory shows Organización alongside each workspace's access, creation date and visibility. The directory starts with All organizations so accessible workspaces from every organization remain visible. Creating a workspace starts with the active organization selected instead, and changing that create-form selection does not narrow the directory. The members panel follows the directory's organization filter; with All organizations selected, it uses the active organization. Use the organization label to distinguish workspaces with the same display name. If organizations share a name, Confluye adds the owner's account name on a muted second line. When that also matches, it adds the organization creation date and, only as a final tie-breaker, a short ID fragment. The sidebar workspace switcher has the same organization selector and shows the active organization when you belong to more than one. The label is shown only for workspaces you can access.

An explicit workspace Owner or Admin can change the display name in Settings → General. The workspace slug, ID, URL, memberships, visibility, API keys, and integrations do not change. Existing bookmarks and automation therefore keep the same address after a rename.

The session tRPC and REST surfaces share the same trimmed 1–120 character name contract and audited rename service. Automated clients can use the confirmed, idempotent workspaces.rename capability; the organization MCP endpoint additionally requires a selected workspaceId. Submitting the current name is a successful no-op and does not write misleading change evidence.

The organization section in General shows the organization display name. A new personal organization starts as Organización de <account name>; if the account name is empty, only the email local part before @ is used. Its Owner can edit and save the name; Admins and Members see it read-only. Renaming refreshes organization labels but does not change the organization slug, workspace URLs, memberships, or access.

A workspace Owner or Admin controls ordinary workspace access. That is not an AIMS role. The AIMS accountable System Owner is a separate governance assignment made in Approvals; the same person may hold both or neither. See AI governance.

General: members

General is Access. An organization Owner or Admin invites external people from the Organization page. After the invitee accepts and joins the organization, a workspace Owner or Admin opens the Organization member selector under Add a member, searches inside its dropdown, selects the person and workspace role, then chooses Assign member. The selector is paginated and includes only people who do not already have an explicit assignment to this workspace. Choose Member, Admin, or Viewer (not Owner). Workspace ownership transfers use the dedicated Owner-only action and are serialized so a workspace cannot lose its final Owner.

Organization Owners and Admins can also cancel a pending organization invitation on the Organization page: choose Cancel beside the invited email and confirm Cancel invitation. The row disappears and its email link stops working. A failure leaves the invitation visible with an inline error. Any pending invitation can be canceled, including one whose link has already expired.

The Workspace visibility control sits above member assignment; Assigned members lists the workspace's explicit role assignments below it. Accepting an organization invitation automatically adds the member to every shared workspace as a Member. Creating a shared workspace likewise grants current organization members Member access, which allows editing workspace content and running automations. Promotion to Admin is only for administrative actions such as managing members and settings. Private workspaces otherwise stay hidden until explicitly assigned. When the certified derived-access rollout (FLUXUS_WORKSPACE_DERIVED_ACCESS) is enabled, organization Owners and Admins receive read-only oversight of private workspaces without edit, execution, or member-directory access. Use the add-member selector only to grant new access. To promote a shared-workspace Member, change that existing assignment's role in Assigned members. Changing a workspace or organization name never changes access.

Once the initial shared-workspace grant is complete, signing in again respects deliberately removed workspace assignments.

POST /api/v1/workspace-members requires a workspace API key with Admin or Owner authority. Invalid assignment roles return 400 "Member role must be admin, member, or viewer.". The email must already be an active member of the workspace organization; otherwise the API returns 409 with organization_membership_required and creates no invitation or assignment.

GET /api/v1/workspace-members lists explicit assignments as { members[] } (role, status, presence, lastActiveAt). A workspace Admin or Owner key is required.

GET /api/v1/workspaces returns { workspace } for the key's workspace.

Preferences

Loading content

Workspace sections show skeleton placeholders while their lists, tables, forms or details load. The placeholders are replaced by the fetched content and respect your browser's reduced-motion preference. If loading fails, use the displayed error and recovery controls to retry. Actions such as saving or running a workflow keep their own progress indicators.

Operation notifications

Operation confirmations use temporary notifications with the workspace's colors. They disappear after six seconds and have a Dismiss notification button (Descartar aviso in Spanish). Repeating a connection check, repository operation, MCP discovery, mailbox operation or CSV import replaces its previous notice. Editable form errors offer Dismiss alert (Descartar alerta); closing an error clears the displayed feedback, and retrying a failed action can show a new error. Closing a notification does not undo an operation or change permissions. Execution diagnostics, deployment blockers and other ongoing status information remain available in their own panels. AIMS workflow actions, setup, role assignments and approval-policy changes use these same operation notifications. MFA and stale-review recovery keep their inline controls and entries.

Interface language

Use Language in the sidebar or Interface language in Settings → General to switch between English and Español. The interface updates immediately without discarding open drafts. The choice is remembered in this browser, including after a reload or a workspace change; it does not change another member's language. Workspace names, documents, chat messages, and workflow content remain in their original language. Navigation and the main interface controls have Spanish translations; specialized controls and existing tutorial text without a translation fall back to English. Public documentation keeps its original language.

Workspace preferences

Canvas/theme/chat provider preferences: session GET|PATCH /api/settings/preferences and v1 GET|PATCH /api/v1/settings/preferences. Patchable fields include theme, autoConnectOnDrop, canvasErrorNotifications, canvasViewport, snapToGrid, showCanvasControls, allowTelemetry, chatAiProvider, chatAiModel, liveTransports, workflow/chat folders and assignments.

For Codex and Claude Code chat, the selected model and reasoning/Fast settings are forwarded to the connected CLI, including models newer than the bundled default list. An invalid selection reports an error rather than silently choosing a different model. Refresh the connected provider's models and select an available entry when a model is unavailable. This chat preference is separate from governed workflow-node model evidence.

Workspace palette

In Settings → General, choose Customize colors to build a shared workspace palette. Edit the light and dark palettes independently: page background, panel background, secondary surface, primary text, secondary text, borders, accent, and accent hover. Each color must be a six-digit hex value such as #2563eb.

Changes stay in the dialog's preview until you choose Save and apply. Cancel leaves the saved palette unchanged. Restore default colors prepares a reset; save it to remove the custom palette. If saving fails, your draft remains available for retry. The palette applies to this workspace for all members, including dialogs and other portal surfaces. The existing System, Light, or Dark selection stays unchanged; System follows the browser's appearance.

The preview reports contrast ratios for primary text on panels, secondary text on panels, and white button text on the accent. Low ratios are warnings about those pairs, not an accessibility certification of the entire interface. Review the screens your team uses before choosing colors.

Reading a palette requires workspace read access; saving or resetting requires workspace write access and records an audit event. Agents use themes.get and the confirmed, idempotent themes.update operations through Platform MCP. This editor manages one palette per workspace; it does not create personal or organization-wide theme libraries.

Claude Code and Codex subscription connections

In Settings → Integrations → Claude Code and Codex, select Login for the provider, finish sign-in in the browser, then select Check. An authorized workspace member connects one subscription credential for the workspace; chat and workflow workers restore that same saved connection into their own runtime homes. Claude Code restores the shared credential even when a service already has files from an older or failed login. You do not need a separate login for each service.

Connected reflects the settings service's latest connection check; it does not prove that a later model request will succeed. If the chat reports a sign-in failure, follow its recovery steps and retry the message after checking the connection. Rate limits and timeouts do not require a new login. See Command Center recovery.

v1 preferences accept any valid API key (not workspace-scope-only). Session PATCH uses workspaceId from the body or the session.

agentRuntimeRollout is not on the preferences PATCH allow-list. Use session GET|PATCH /api/settings/agent-rollout as Admin/Owner. Surfaces: agentNode, copilot, home. Modes: legacy, shadow, v2, or null to revert to env default. Audited as agent_runtime.rollout.update.

Git review repository

Open Settings → Workflow reviews to configure the GitHub repository, base branch, artifact path and workspace GitHub credential. If no GitHub credential exists, Connect GitHub credential lets you paste and link one in this section. The integration catalog and its search remain in Integrations.

Require merged PRs for previews is editable by workspace Admins and Owners. Off allows a preview to test an exact commit from an open PR; on requires a verified merge for previews as well. Production always requires merge when a repository is linked. AIMS approval is required separately for every live destination. Policy changes apply to the next publication and do not stop existing deployments.

The API field is policy.requirePreviewMerge (boolean). Omitting policy while updating repository fields preserves the stored policy.

Session GET|PATCH|DELETE /api/settings/repository (tRPC settings.getRepository / upsertRepository / deleteRepository). GET: any member. PATCH/DELETE: Admin or Owner. PATCH requires owner and repo; defaults defaultBranch: main, pathPrefix: workflows. Response includes repositoryAccess (verified / warning / failed) from a GitHub list-files probe. There is no v1 repository route.

Authorization summary

Most Settings reads need only membership. Stricter roles: repository write (Admin/Owner), agent rollout (Admin/Owner), workspace member discovery and assignment (workspace Admin/Owner), v1 member assignment (Admin/Owner), observability retention POST (Owner — see recovery).

Provider connection verification

Use Test in Integrations to verify that the saved workspace credential can authenticate. Before a release, also execute the provider's workflow block against a dedicated test resource: credential testing alone does not prove node mapping, live controls, runtime scopes, or result handling. Read probes should run first. External writes require an explicit confirmation and a cleanup plan.

The separate S3-compatible first page account is an exception: genericS3ManualSigV4V1 saves an explicit custom HTTPS origin, region, addressing mode and supplied keys without contacting the endpoint. Named fields and Advanced JSON both save encrypted, write-only material; reopening never fills stored secrets, and updating requires a complete replacement. Test and bulk Test are unavailable, and saving leaves access unverified. Its workflow action requests only one bounded first page after the three literal acknowledgments and separate Human approval; it performs no discovery or continuation. See the block reference.

OpenAI voice transcription uses the workspace AI provider and gpt-4o-mini-transcribe. Audio remains bounded by the upload and duration limits; Confluye records duration/model usage, not audio bytes or transcript content. Individual Google Slides REST connections request presentations and drive.file; their content operations can access existing presentations by ID. Official Google MCP connections have separate service profiles and are configured under MCP Servers.

Voice recordings are limited to 10 MB and 60 seconds. Duration is verified from the uploaded audio, including fragmented MP4 recordings, and provider-reported duration can increase the metered amount. A provider request that completes can still incur usage when it returns an empty transcript or the recording fails the final duration check.

Meeting reminder templates

For calendar reminders, use Home → Workflow templates → Productivity → Multi-calendar meeting reminders. Connect multiple Google Calendar or Outlook accounts through the regular integration cards, select one credential per source block, and replace the notification block with your preferred channel. The template uses the regular workflow editor, schedules and execution history. Calendar data follows workspace access rather than a private personal agenda. See the setup and delivery limits.

Personal browser connections

Manage my browser connections opens the personal connection screen. The deployment administrator must configure the browser worker first. If you open the screen directly, choose your workspace before adding a connection. Each connection can use a different website, including Centris. Enter a connection name, HTTPS login page, HTTPS account page, and exact text that uniquely identifies your signed-in account on that page, such as your displayed email. Do not use text also shown before login. Add any extra domains needed when the agent reuses the session, such as single sign-on or page resources. During human login the private browser can follow any public HTTPS website, including SSO and MFA redirects. Private networks and non-HTTPS destinations remain blocked. Saving keeps only cookies and storage belonging to the configured site domains; the agent then uses that domain allowlist. Extra domains are used only for login and page resources: agents read pages only on the account page's domain, and a read blocks any request that could change the account, such as a form submission.

Add the connection, choose Open browser, and complete login and MFA in the private browser viewer. The browser fills the available screen and adapts to mobile width. Tap or click a website field and type directly; use your keyboard, mouse wheel or touch swipes normally. Enter credentials for that website manually; the viewer does not provide Confluye password autofill. The browser bar shows the current domain and provides navigation and Refresh browser after an external MFA step. Save and verify closes that browser and checks the same account in a new session. Both persistent cookies and session cookies are restored; their expiration is not extended. Only a successful check marks the connection ready. Sessions can expire and require another interactive login. If Save and verify reports an error after the browser closed, the connection stays in saving; choose Save and verify again to finish.

Connections belong to their creator in the selected workspace. Create login link issues a private, single-use link that expires in 15 minutes for a new, disconnected connection. While the recipient's temporary access is active, you cannot view or control that browser. Send it only to the intended account holder: possession of this link grants access to that connection's login screen, without registering or signing in to Confluye. It never grants access to the rest of the app. The guest uses the same direct browser controls, without a separate text-entry form or keyboard buttons. For a temporary loading or viewport-sizing error, choose Retry; use the address bar or Back after a failed navigation. A slow preview capture keeps an open login available for Retry without sending the login again. If the page says the browser stopped and needs recovery, ask the sender for a new connection link; this is a browser failure even when the invitation has not expired. Typing is grouped into short batches instead of sending a request for each letter. Clicking, pressing Enter or navigating sends any earlier text first. Saving waits until typing, other input and the current preview update finish in both owner and guest viewers. The guest preview updates every two seconds for up to 20 seconds after opening, interacting or choosing Refresh browser, so a delayed navigation can finish without another click. Updates pause during input and stop after a snapshot error, when idle or when the viewer closes. If external MFA takes longer, choose Refresh browser to resume.

Opening or previewing the link does not consume it. The recipient selects Connect my account, then Open secure browser, and signs in to the website. Save and allow access verifies the saved session, enables agent reading, closes the viewer and ends the temporary access. The recipient can close the page and return to the conversation. Temporary access lasts at most 15 minutes after redemption. If saving reports a temporary error, retry Save and allow access. Reloading the page recovers a completed confirmation without reopening the browser. If cancellation fails, retry Cancel connection; browser control is already blocked while cleanup is pending. After the link session expires, ask the sender to finish cleanup. A viewer that reaches its ten-minute limit can be opened again to continue login. Issuing a replacement link before the browser opens invalidates the previous link. Links cannot expose an existing account, even if its session has expired. Create a new connection to request another person's login. Old invitations issued before this feature require a replacement link.

Disconnect and delete profile immediately blocks reuse and requests provider cleanup. If a browser operation is still running, the status remains revoking until cleanup succeeds; retry the button or contact the administrator. Recovery required also needs administrator attention. Your saved login stays on the dedicated Playwright worker in your deployment; Confluye stores encrypted references and site configuration. The saved cookies, local storage and site data are encrypted with a worker-only key. Encrypting the worker volume at rest is still recommended. Cookies are never returned to the agent.

Agents and workflow blocks can use the same connection at the same time; each action runs in its own temporary browser. Some websites sign you out when one session is used from several places at once. If that happens, sign in again and avoid running simultaneous actions for that site.

The preview Web browser workflow blocks offer Verify saved access, Get connection status and Read authenticated page. Select your connection, enter a page URL when reading a page, save the block, then use Authorize this block and save again. Authorization is limited to the same owner, workspace, workflow and block. Revoke block access disables subsequent use. Copying or exporting a workflow clears these account bindings. Existing Centris verification blocks remain compatible.

For Codex CLI and Claude Code workflow nodes, select a connection under Playwright MCP · saved browser, save, choose Authorize browser control, then save again. This separate control grant lets the agent navigate, click and fill forms, including submitting changes, through the existing MCP attachment mechanism. Once browser control is authorized, these browser tools can run unattended. This grant does not authorize tools on other MCP servers, and revoking it stops subsequent browser calls. Instruction skills remain available. Other selected MCP tools remain available unless the node also selects product connection tools; a node with product tools may attach only those tools and this run's saved-browser session. Each node run gets its own temporary browser from your saved login, which is saved and closed automatically; other runs and agent reads can use the same connection at the same time, so avoid running simultaneous changes to the same account. Revoke block access disables subsequent calls. Copy/export clears the binding. See MCP browser control.

For a product connection, a Codex CLI or Claude Code node needs workflow trust. Confluye gives the CLI a temporary loopback MCP door with only the node's selected product tools and signs requests with the run's saved identity. The CLI never receives the signing secret or signed run context; Confluye sends signed headers directly to the product. A node with product tools cannot attach another MCP server. An untrusted connection or mixed attachments block validation, publication and run admission. See the product connections reference for trust and refusal behavior.

For conversational agent access after connecting through the owner screen, choose Allow my agent to read. The recipient's Save and allow access already grants this permission in the private-link flow. Your agent can list your connections and read visible page text and links on each connection's account site using browser.list and browser.read. Remove agent read access stops subsequent agent reads without deleting the connection or changing a separately authorized workflow block. Other users cannot reuse your session. Page content is untrusted website data. These conversational tools support navigation and reading; form interaction requires the separate CLI node control grant above. Document-signing integrations and automatic Telegram account linking are not included.

End users and memory

When the deployment runs with FLUXUS_END_USERS=true, Integrations also shows End users and memory. Its Manage end users and memory button opens the console inside this workspace's app layout, preserving its navigation and theme: the external customers your product serves, the memory resources your workflows share per customer, and the websites and site notes those customers use. Workspace Admins and Owners with explicit membership can read and change customer content and delete a memory resource; every read is audited. The card is hidden while the flag is off. See End users and memory.

Ghost Content and Ghost Admin have separate live-execution controls and native saved accounts. Saving either account sends no provider request or JWT and remains unverified; Test connection is unavailable. Select the correct profile in each block, review personal-data and write-effect acknowledgments, and confirm each invocation separately. See Ghost setup and limits, including owner notifications during API version negotiation and uncertain write outcomes.

Next steps

Gotify has a separate workspace live-execution control and native gotifyApi account. Save sends no provider request and remains unverified; Test connection is unavailable. Select the exact native account, review initially unchecked privacy and write-effect acknowledgments, and confirm each invocation separately. Application tokens create notifications; client tokens read or delete messages. See Gotify setup and limits, including uncertain write outcomes and the one-page metadata limit.

Hunter has a separate workspace live-execution control and native hunterApi account. Save leaves authentication unverified; Test is unavailable. Select the exact account and review personal data, plus credits and automatic lead changes for finding or verifying addresses. See Hunter setup.

Home Assistant has its own live-execution control for the five native reads. Its explicit workspace value overrides the environment fallback; enabling it does not validate credentials or replace the initially unchecked personal-data acknowledgment and separate invocation confirmation. See Home Assistant setup.

Humantic AI has its own workspace live-execution control. Its explicit workspace value overrides the environment fallback and does not validate the selected native account or replace the personal-data/processing-credit acknowledgments and separate confirmation. Creation starts processing even though the request uses GET. See Humantic AI setup.

Iterable has its own workspace live-execution control. Its explicit value overrides the environment fallback and does not validate an account or replace personal-data, write-effect, channel-unsubscription acknowledgments or independent invocation confirmation. See Iterable setup.

HighLevel has a separate workspace live-execution control whose explicit value overrides WORKFLOW_LIVE_HIGHLEVEL. Enabling it does not validate the selected PIT or replace personal-data, write-effect, upsert acknowledgments or independent invocation confirmation. See HighLevel setup.

Keap has its own workspace live-execution control. An explicit setting overrides WORKFLOW_LIVE_KEAP; enabling it does not validate the saved token or replace the unchecked personal-data acknowledgment and separate invocation confirmation. See Keap setup.

Invoice Ninja has a separate workspace live-execution setting that overrides WORKFLOW_LIVE_INVOICENINJA when explicitly set. It does not verify the selected hosted-v5 token or replace the unchecked privacy/effect acknowledgments and separate invocation confirmation. See Invoice Ninja setup.

LinkedIn has its own workspace live-execution control. An explicit value overrides WORKFLOW_LIVE_LINKEDIN; enabling it does not verify the saved OAuth token, app access or author role and does not replace the unchecked publication acknowledgment or independent invocation confirmation. See LinkedIn setup.

Mandrill has a separate workspace live-execution control. An explicit value overrides WORKFLOW_LIVE_MANDRILL; enabling it does not verify a saved key, live/test status, rights, billing or delivery and does not replace the three unchecked acknowledgments or independent send confirmation. See Mandrill setup.

LingvaNex and Magento have separate workspace live-execution controls overriding WORKFLOW_LIVE_LINGVANEX and WORKFLOW_LIVE_MAGENTO2. Enabling either does not verify the selected account or replace initially unchecked operation acknowledgments and independent confirmation. Save remains local/unverified and Test is unavailable. Review translation text/charges and merchant privacy/account/financial/fulfillment effects separately. See LingvaNex and Magento.

Home Assistant actions, Matrix, Mailcheck, Monica Classic4.x, Mindee InvoiceV4 and MessageBird have separate workspace live controls. Explicit settings override WORKFLOW_LIVE_HOME_ASSISTANT_ACTIONS, WORKFLOW_LIVE_MATRIX, WORKFLOW_LIVE_MAILCHECK, WORKFLOW_LIVE_MONICA_CLASSIC4X, WORKFLOW_LIVE_MINDEE and WORKFLOW_LIVE_MESSAGEBIRD respectively. Enabling them verifies neither saved accounts nor rights and does not replace unchecked applicable policies or independent invocation confirmation. Home Assistant actions uses a separate account from the existing five reads. See setup and47-operation boundary.

Mocean adds an explicit workspace live control overriding WORKFLOW_LIVE_MOCEAN_V2. Mindee's existing WORKFLOW_LIVE_MINDEE control covers separately selected InvoiceV4 and ReceiptV5 accounts; Monica Calls use the existing WORKFLOW_LIVE_MONICA_CLASSIC4X control. Enabling execution verifies neither saved credentials nor permissions and does not replace unchecked applicable policies or independent confirmation. See setup.

Mistral adds WORKFLOW_LIVE_MISTRAL_AI and global Microsoft Teams Graph v1 reads add WORKFLOW_LIVE_MICROSOFT_TEAMS_GRAPH_V1. Explicit workspace settings override operator switches. Enablement proves neither saved credentials nor account eligibility and does not replace exact account selection, initially unchecked applicable privacy/paid-processing controls or separate confirmation. See setup.

The existing WORKFLOW_LIVE_MONICA_CLASSIC4X and WORKFLOW_LIVE_MICROSOFT_TEAMS_GRAPH_V1 controls also govern Activities5 and Teams WRITE9. Enablement does not grant data rights or replace exact saved-account selection, initially unchecked relationship/data/change/overwrite/deletion policies or independent invocation confirmation. See the fourteen field and response contracts.

Perplexity and OpenThesaurus use separate workspace live controls WORKFLOW_LIVE_PERPLEXITY and WORKFLOW_LIVE_OPENTHESAURUS. Perplexity additionally requires an exact selected native account; OpenThesaurus requires all credential references/bindings unset and has no account dialog. Enablement alone does not grant execution, replace unchecked acknowledgments or create human confirmation. Organization membership alone is insufficient. See the two field, response and privacy contracts.

Outlook token reads add WORKFLOW_LIVE_MICROSOFT_OUTLOOK_GRAPH_V1; Monica conversations/contacts use existing WORKFLOW_LIVE_MONICA_CLASSIC4X. Saved explicit false overrides a true environment switch and current-live state is rechecked for these23 actions after callback confirmation and preparation. Enablement does not grant mailbox/relationship authority or replace selected native accounts, initially unchecked acknowledgments or separate confirmation. See the23 field and response contracts.

The same native Outlook token account now also supports sixteen writes, for28 finite Outlook operations. Use the existing workspace live control and exact selected native account; enablement does not verify mailbox rights or replace initially unchecked applicable acknowledgments plus independent confirmation. Draft send uses stored undisplayed recipients/content/attachments, and general message deletion is not draft-only. See Outlook token writes.

urlscan adds WORKFLOW_LIVE_URLSCAN_V1 and Marketstack adds WORKFLOW_LIVE_MARKETSTACK_V2. Explicit workspace false overrides a true environment switch. Exact saved native accounts and current live state are checked after both dispatch callbacks; enablement does not verify rights or create consent. urlscan submission needs an initially unchecked disclosure plus independent confirmation; scan.get has neither requirement. All three Marketstack reads need an unchecked query/quota acknowledgment and separate confirmation. See five finite actions.

npm metadata execution control

Enable npm-public-v1 only for the intended workspace; an explicit workspace false refuses dispatch. WORKFLOW_LIVE_NPM_PUBLIC_V1 is the operator setting when no workspace override exists. No account needs saving. The selected package/version or search query and request metadata go to the public registry, and execution requires separate confirmation. See the npm field, response, privacy and timing contract.

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 and KoBo accounts

Save an existing Telegram public Bot API token or a KoBo token bound to its exact Global, EU or public self-hosted origin. Named fields and Advanced JSON store encrypted credentials locally and reopen as write-only/unverified; Save makes no HTTP or DNS request. Neither profile offers credential Test, OAuth, refresh or option discovery. Enable its workspace live transport and select the exact saved account on the block. Saving an account grants no request approval. See Telegram and KoBo for explicit migration, request approval and capability limits.

AWS Comprehend, SNS and SES accounts

Save existing manual SigV4 credentials using awsComprehendManualSigV4V1, awsSnsManualSigV4V1 or awsSesManualSigV4V1. Choose a supported commercial region and static keys or a supplied temporary session with explicit UTC expiry. Named fields and Advanced JSON use the same encrypted, write-only account. Save and reopen send no provider request and leave authentication unverified; Test and bulk connection testing are unavailable for these profiles. No STS, role assumption, discovery or refresh occurs. The declared account ID is not verified identity. Exact raw response bytes are application-encrypted in WorkspaceFiles; the configured scanner inspects ciphertext. Operation data can remain in workflow history and reach downstream AI. Review the service-specific limits and retention acknowledgement in Integrations and credentials.

HaloPSA accounts

Use haloExistingBearerV1 for an existing manual token, exact public HTTPS origin, explicit API base path and operator-supplied expiry. Named fields and Advanced JSON keep tokens and optional user passwords write-only. Save/reopen is local and unverified; Test and token acquisition are unavailable. Follow HaloPSA setup and protected results.

GoToWebinar accounts

Select gotoWebinarExistingBearerV1, then enter an existing bearer, its actual UTC expiry, and decimal-string account/organizer keys. Named fields and Advanced JSON are equivalent encrypted, write-only saves. Save and reopen remain unverified; they send no provider HTTP/DNS. Test, OAuth acquisition/refresh and account discovery are unavailable, and bulk tests skip the profile. Existing app/account entitlement and collaboration permissions must be verified separately. See GoToWebinar setup and response retention.

FileMaker accounts

Choose filemakerManualSessionV1, explicitly choose Data API v1, and enter the exact public origin, database, existing session token and actual operator asserted UTC expiry. Named fields and Advanced JSON share encrypted write-only Save/update/reopen. These local operations send no provider request and do not verify session liveness, permissions or schema compatibility. Test and bulk checks do not contact FileMaker; login/logout/refresh and metadata loaders are unavailable. An idle session may expire sooner than the supplied timestamp. Repair the selected account explicitly rather than retrying an uncertain action. See exact profile fields.

Bitwarden existing organization bearer session

Choose bitwardenOrganizationBearerSessionV1 and explicitly enter US, EU or public self-host origin profile, exact origin, organization UUID, operator session nonce, existing bearer and actual UTC expiry. Self-host additionally requires its version. Deliberately check the migration/session acknowledgments and, for EU/self-host, regional migration. The application does not acquire or refresh this token or infer authority from its contents.

Named fields and Advanced JSON share strict encrypted Save/update/reopen. Tokens remain write-only, the account stays unverified and these local operations send no provider HTTP/DNS. Test is unavailable and bulk checks skip. Replace an expired or changed account explicitly; prior approvals cannot authorize a replacement session. See profile requirements.

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.

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 account

Choose awsAcmManualSigV4V1 with the exact aws-acm-manual-sigv4-v1 profile, supported pinned region, declared account and manual-static or supplied manual-temporary session plus actual expiry. Named and Advanced JSON Save/update/reopen use the existing encrypted write-only account and send zero provider HTTP/DNS. The account stays unverified; Test, bulk Test, STS, role/system identity, custom endpoints and refresh are unavailable. See profile and retention requirements.

DynamoDB typed items

Choose the DynamoDB manual profile with named fields or Advanced JSON. Save/update/reopen is encrypted, write-only, local and unverified with zero provider HTTP/DNS. Test and bulk Test are unavailable. Supply actual temporary-token expiry rather than a guessed or refreshed session. See typed requests, examples and limits.

AWS ELB ordinary actions

AWS ELB uses a separate encrypted awsElbManualSigV4V1 account and workspace aws-elb live policy. Save/update/reopen is local, unverified and zero HTTP/DNS; Test and bulk Test are unavailable. Saving grants no request approval. Temporary tokens require actual UTC expiry, and execution requires more than ninety seconds remaining. Select the exact manual account on each block. 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 account

Choose awsCognitoManualSigV4V1 and the exact aws-cognito-manual-sigv4-v1 manual profile. Supported regions are us-east-1, us-east-2 and us-west-2; temporary mode requires a supplied token and actual UTC expiry, with more than ninety seconds remaining at dispatch. Named and Advanced JSON Save/update/reopen encrypt write-only secrets locally, remain unverified and send zero provider HTTP/DNS. Test/bulk Test, OAuth/login, STS/role/system credentials, acquisition, refresh and discovery are unavailable. See admin scope and consequences.

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.

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

Use odooApi with an explicit public HTTPS origin, database, username and existing password/RPC API key of at least eight characters. Named and Advanced JSON Save/update/reopen are encrypted, write-only, local and unverified, with zero provider HTTP/DNS. Test and bulk Test are unsupported. The saved account grants no execution approval.

Deployment/version, Custom plan, model, ACL and public reachability are unverified. Local provider/Prisma facts are simulated; real provider/native/DB/Review organization and workspace behavior is untested. See the complete bounded protocol and limitations.

QuickBooks manual bill-read account

Choose QuickBooks and quickbooksBillManualBearerV1 in Integrations. Supply an existing access token, external decimal-string company realm, explicit production/sandbox environment, accounting scope and expiry metadata. Named fields and Advanced JSON save encrypted, write-only, local/unverified accounts with no provider HTTP/DNS. Reopen, re-enter the entire replacement profile and choose Update account to keep the same ID; the existing token is never displayed. Test, OAuth acquisition/refresh and automatic legacy credential conversion are unavailable. Realm is distinct from Confluye organization/workspace; Save does not verify token grants, scope, expiry, company access, app approval or quota.

Only a separately approved one-bill read can execute. Review its initially unchecked literal financial-data persistence acknowledgment because the Bill and raw response JSON enter ordinary history/downstream results; parsed decimals are not exact financial arithmetic. See QuickBooks setup and restrictions. Current organization AND workspace authority remains required. Real provider, native network, durable restart and Review Help walkthroughs remain untested.

AWS Rekognition DetectLabels (preview)

A saved Rekognition account remains unverified. Enabling the workspace live preference does not validate AWS identity, image access or retention policy, and does not replace the deployment setting or Human approval. Turning the preference off or changing current approval material prevents the pending request from being sent. Temporary accounts also need at least ninety seconds of remaining validity at dispatch.

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

Una cuenta Rekognition guardada permanece sin verificar. Activar la preferencia de ejecución real del workspace no valida identidad AWS, acceso a la imagen ni retención, y no sustituye la configuración del despliegue ni la aprobación humana. Desactivar la preferencia o cambiar los datos de aprobación vigentes impide enviar la solicitud pendiente. Las cuentas temporales también necesitan al menos noventa segundos de vigencia al enviar.

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)

The Textract live preference uses the exact provider key aws-textract. Saving that preference preserves the existing live keys, including Rekognition. Enabling it grants no AWS account access and does not bypass Human approval. Both the deployment and workspace gates are required at execution, including after a paused run resumes.

The original actor must still have current access to both the organization and workspace. The runtime rechecks the saved encrypted account revision, source, workflow version and exact ready file content. Disabling live access, changing the account or file, revoking access or reusing a consumed approval refuses dispatch. These checks do not make database and network actions atomic.

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

La preferencia de ejecución real usa la clave exacta aws-textract. Guardarla conserva las demás claves, incluida Rekognition. Activarla no concede acceso a una cuenta AWS ni evita la aprobación Human. Se exigen las dos autorizaciones de despliegue y workspace, también al reanudar una ejecución pausada.

El actor original debe conservar acceso tanto a la organización como al workspace. Se comprueban otra vez la revisión de la cuenta cifrada, el origen, la versión del flujo y el contenido exacto del archivo listo. Desactivar el acceso real, cambiar cuenta o archivo, revocar acceso o reutilizar una aprobación consumida impide el envío. Estas comprobaciones no hacen atómicas la base de datos y la red.

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)

In Settings → Integrations, the Microsoft SQL manual account supports encrypted Save and full replacement through Update account. Choose credential type microsoftSqlManualTds74TlsV1 and supply all six fields: profileId: "mssql-manual-tds74-tls-v1", host, numeric port, database, username, password. Reopened fields are blank. Re-enter the entire account to update it; omitting a field or the replacement secret is refused by Settings. Existing credential permissions and sensitive-action verification still apply.

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. See account setup.

Español: cuenta manual de Microsoft SQL (solo almacenamiento)

En Configuración → Integraciones, la cuenta manual de Microsoft SQL permite guardar con cifrado y reemplazar todo el perfil mediante Actualizar cuenta. Elige el tipo microsoftSqlManualTds74TlsV1 y aporta los seis campos: profileId: "mssql-manual-tds74-tls-v1", host, port numérico, database, username, password. Al reabrir, los campos quedan vacíos. Vuelve a introducir la cuenta completa para actualizarla; Configuración rechaza la omisión de un campo o del secreto de reemplazo. Siguen vigentes los permisos de credenciales y la verificación de acciones sensibles.

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. Consulta la configuración de la cuenta.