Permanently delete a task. Comments and attachments are removed automatically. The taskId is the internal document ID from list_tasks — NOT the public task number (e.g. SCD-408).
Connector
Destructive
OAuth669,417 tools. Updated 2026-10-17 10:05
"A server for task number 2 automation" matching MCP tools:
- Delete an automation and its run history. Cannot be undone. A live automation is refused with automation_live: disable it first, or pass force: true after the person confirms, because deleting it stops what is going out to people.ConnectorDestructiveOAuth
- Create a new automation DEFINITION for the authenticated user. The new automation is always created INACTIVE (a definition only) — it does NOT run, send, or activate. Use when the user wants to define a new automation by name. Do NOT use to activate, run, or wire the automation's step graph — activation is a separate, later capability and is not available here. Do NOT use to edit an existing automation — use mcp_crm_automations_update. The automation's status cannot be set via this tool. Tenant is bound to the session and cannot be changed via arguments. Automation-level RULES (all optional, and all independent of the step graph — use mcp_crm_automations_set_steps for that): entryConditions / exitConditions / restartConditions are arrays of {type, segmentId, operator, contactFilters, campaignIds, linkUrl} where type is one of added_to_lists, removed_from_lists, contact_in_segment, contact_matches_filters, email_opened, email_clicked, email_unsubscribed; a membership type needs segmentId, contact_matches_filters needs contactFilters, and the email types take optional campaignIds (plus linkUrl on email_clicked). operator is MATCHES or DOES_NOT_MATCH. Pass an EMPTY array to clear a slot; omit the key to leave it unchanged. reentryPolicy is once (a contact runs this automation at most once) or every_time, with reentryWaitMinutes as the cool-down before a contact may re-enter. onDeactivate is continue (in-flight contacts finish) or cancel (in-flight contacts are cancelled when a human deactivates the automation). Returns the created automation projection (id, name, status, rules, version).ConnectorOAuth
- Edit an existing active task identified by its UUID or task reference. Prefill a native form with its current title, description, due date and project workflow status, then save only accepted changes. Cancelling makes no change. Requires a host supporting MRTR form elicitation; otherwise use item_update. Accepted task changes may send configured notifications, webhooks and automation actions.ConnectorDestructiveOAuth
- Creates an automation on a perspective. Triggers: per_interview (fires on every completed conversation) or scheduled (daily/weekly, with digest, invite, or insights purpose). Actions: internal insight discovery, webhook, email, or connected provider-backed integrations such as Slack, HubSpot, Gmail, Google Docs, Notion, and Confluence. Execution modes: direct (fast, deterministic, webhook-only) or agent (LLM-powered, required for insights and delivery channels). Behavior: - Each call creates a new automation — even if name/config matches an existing one. - Once enabled, the automation starts firing on real events: per_interview sends on every completed conversation going forward; scheduled sends a real message on the configured cadence (daily/weekly). - For HubSpot, the workspace's HubSpot connection is required — errors with "Could not resolve HubSpot portal ID — please reconnect HubSpot" if not connected. - Webhook channels: do NOT ask the user for the endpoint URL or credentials — neither is accepted through this tool. The automation is created disabled and the response includes configure_url, a web app page where the user sets the URL (and an authentication header if needed). Share that link and ask the user to reply "Done" after saving, then enable the automation via automation_update. - Insight automations use kind "insights", execution_mode "agent", omit channel, and use scheduled purpose "insights" when scheduled. - Errors when the perspective is not found or you do not have access. When to use this tool: - The user wants ongoing notifications on every completed conversation (per_interview). - Building a daily/weekly digest delivered to Slack, email, HubSpot, or a webhook (scheduled). - Running scheduled insight automation which creates insights without external delivery. When NOT to use this tool: - Trying a one-off send before going live — create the automation, then use automation_test (use override_email on email channels to avoid hitting real recipients). - Editing or toggling an existing automation — use automation_update. - Connecting Slack or HubSpot — use integration_manage first; the provider must be connected before slack/hubspot channels work. Example — per-conversation Slack notify (resolve the channel with slack_channel_resolve first, then pass it as resource_id): ``` { "perspective_id": "...", "automation": { "name": "Notify Slack", "trigger": { "type": "per_interview" }, "execution_mode": "agent", "channel": { "type": "composio", "delivery_config": { "provider": "slackbot", "tool_slug": "SLACKBOT_SEND_MESSAGE", "resource_id": "C0123ABCD", "resource_name": "#research" } } } } ``` resource_id is the Slack channel ID or name. The channel is re-verified live on create; an unresolvable channel is rejected. Typical flow: 1. integration_manage (operation: "list"/"connect") → ensure Slack / HubSpot is connected (only needed for those channels) 2. For Slack: slack_channel_search / slack_channel_resolve → find/verify the channel to use as resource_id 3. automation_create → create the automation 4. automation_test (with overrides) → verify delivery before relying on itConnectorOAuth
- Mark an existing task as completed or reopen it. Use this instead of create_task when the task already exists and you only need to change its completion state. The taskId is the internal document ID from list_tasks — NOT the public task number (e.g. SCD-408). To resolve a public number to an internal ID, call list_tasks with workspaceId + number first.ConnectorOAuth
Matching MCP Servers
- AlicenseNot gradedqualityBmaintenanceGenerates a unique number derived from the International Space Station's current orbital location, useful for space-derived identifiers or deterministic random seeds.31 npmMIT
- AlicenseNot gradedqualityBmaintenanceProvides a random number aged 24 hours, offering a smooth, aged number for any purpose.33 npmMIT
Matching MCP Connectors
- Parallel Task MCPOAuth
An MCP server for deep research or task groups
Programmatic headless browser sessions: create, interact, browse, and extract - rendered in a real
- The people in one automation, newest enrolment first: each row has the email, status, current_step, next_due_at and, for a cancelled one, cancel_reason. It answers 'is this person still on the sequence?', 'who is waiting on step 2?' and 'why did this sequence stop for them?', which find_contact and list_emails do not. status narrows to one state: 'active' (still going, including people held while the automation is paused), 'completed', 'failed', or 'canceled' for everyone who left early, where cancel_reason says why (unsubscribed, unsubscribed_from_topic, replied, suppressed, exit_tag, required_tag_missing, manual). For the counts alone, list_automations already carries them per status. Any other status is refused with 422 invalid_request, and an unknown automation_id answers 404 not_found. At most 100 per call; while has_more is true, next_cursor passed back as cursor with the same status returns the next page.ConnectorNo auth
- Start one of the user's automations now, the same as "Run now" in the web app. The AI agent then drives the automation's phones on its own; this returns the run at once with status "queued". Poll get-automation-run-tool with its `id` until `status` is done, failed or stopped, then read `outcome`, `result` and `phone_reports`. A run can be refused (the automation is off, a run is already going, the daily limit is reached, no phone is online, or the agent is unavailable); the error says which. A `tiktok_warmup` automation (see list-automations-tool) has no daily limit and starts one session per phone at once, so this returns the first one already "running" alongside `sessions`, the number started, and `skipped_busy_slots`, the phones left alone because another run already drives them; `payload` is ignored, it has no parameters.ConnectorDestructiveOAuth
- Automatically send an approved template to every NEW contact with a tag, e.g. a welcome message whenever a lead tagged "new-lead" is added (by add_or_update_contact, import_contacts, the REST API, or a Zapier/Make integration calling it). Only contacts created AFTER the automation qualify, each at most once: re-importing or tagging existing contacts never triggers it. To message existing contacts, use a broadcast; the response's existing_tagged_contacts says how many already have the tag. Only add contacts who agreed to receive WhatsApp messages. The template must be APPROVED with every {{n}} placeholder mapped (set_template_variables). One automation per tag. delay_minutes (0 to 10080) waits before sending. Returns the automation with live stats and any warnings (e.g. no WhatsApp number connected) that would make sends fail.ConnectorOAuth
- Create a new task by asking the person for its title, description, due date and project workflow status in a native form. An optional projectId selects the project. Save only accepted values; cancelling makes no change. Requires a host supporting MRTR form elicitation; otherwise use item_create. Accepted task creation may send configured notifications, webhooks and automation actions.ConnectorDestructiveOAuth
- Permanently deletes an automation. Pauses any scheduled sends first, then removes the automation. Behavior: - DESTRUCTIVE and irreversible — the automation cannot be recovered. No undo. - Errors when the perspective or automation is not found, or you do not have access. Deleting an already-deleted automation errors as well. - If pausing the scheduled sender fails, the deletion is aborted and you'll get success: false with "Failed to stop running workflow. Please try again." — the automation stays intact in that case. When to use this tool: - The user explicitly asked to remove an automation and confirmed. - Cleaning up a misconfigured automation that automation_test repeatedly fails on. When NOT to use this tool: - The user just wants to pause it temporarily — use automation_update with { enabled: false } instead. - You're not sure which automation_id is correct — confirm via automation_list first.ConnectorDestructiveOAuth
- Creates a task in OmniFocus on this Mac. Needs OmniFocus Pro (the free tier blocks automation). Without `project` the task goes to the inbox. `project` must be the exact name of an existing project: if no project has that name the call fails with an error instead of falling back to the inbox — check the name with list_omnifocus_projects first. Called without confirm it returns a preview; pass confirm=true to create. Returns the task id.ConnectorOAuth
- List tasks in a project, or look up a specific task by its public number. To resolve a public ID like 'TES-1': pass taskPrefix='TES' and number=1 — the prefix locates the correct workspace automatically (do NOT hardcode workspaceId when the prefix belongs to a different workspace). Alternatively pass workspaceId + number when you are certain of the workspace.ConnectorOAuth
- Returns a real LORG COUNCIL-tier contribution with a score breakdown and annotations. Call this after Task 1 and before submitting Task 2 — it shows exactly what a high-scoring contribution looks like and why each dimension scored well.ConnectorNo auth
- Spawns a background LLM agent to handle a task. Checks RAM (needs 500MB+), enforces max 2 concurrent dispatches. Returns PID and log file path.ConnectorNo auth
- Search real student IELTS Writing answers scored by AI examiner. Args: task: IELTS Writing task number — 1 (reports/letters) or 2 (essays). band: whole band to filter on; matches half-bands too (8 -> 8.0-8.5). query: free-text match against the question/prompt. limit: max results (1-50). Each result links to a full page with the question, the student's answer, and criterion-by-criterion examiner feedback. Use get_model_answer for the full content, or compare_question to see the same prompt at other bands.ConnectorNo auth
- Look up the reference chart for one Kua number: its trigram, its east or west life group, and how it classifies all eight compass sectors. A pure reference endpoint with no birth data required, for building a lookup table or a picker. Number 5 is served for completeness and is never a computed result, because it belongs to the centre and has no direction of its own: a man whose formula gives 5 reads Kua 2 and a woman reads Kua 8, and the chart returned for 5 is therefore the Kua 2 chart.ConnectorNo auth
- Reboot an iOS device. DISRUPTIVE and NOT REVERSIBLE: the device is unreachable for roughly a minute, the automation session and the stream go down, and anything in progress on the device is lost. Requires confirm=true — there is no default. Do NOT reach for this to recover from a flaky tool call; try ios_end_session first, which resets the automation session without touching the device. Reboot only when the device itself is wedged and a human has asked for it. No automation session required. The Android counterpart is device_reboot.ConnectorDestructiveOAuth
- Use this when the user wants something done for them on a schedule — every day, on certain weekdays, or every few hours — e.g. 'every Monday at 9, summarise my sales sheet'. Creates a Xenition automation that runs on its own at those times, even when nobody is in this chat. Results appear in Xenition, and anything that would act in a connected app waits for the user's approval. Do not use for a one-off task now (use run_agent).ConnectorNo auth
- Delete a DM automation for good, with its stats and runs; people in the flow stop. To stop it for a while instead, use update_dm_automation with active false.ConnectorDestructiveAPI key
- THE approval inbox for V2 rule automation — the ONLY surface for pending automation approvals (MinMax/AOE/scheduled-task tools are separate systems, not this inbox). status=pending (default) means NOT yet approved/dismissed by a user: pending_approval (manual queue + escalations), pending_ai_review (queued for AI Workforce Review — the Automation Review Analyst seat works this queue at its check-ins), shadow (legacy AI Shadow rows; needs your approve/dismiss) and ai_rejected (overridable). Rows carry approval_mode (manual|workforce_review|auto; legacy ai_review/ai_shadow read as workforce_review), rule owner, needs, undoable, and paginate with total_found/truncated/next_offset. scope=mine (default) shows only the calling user's rules; scope=all_users shows every user's. READ-ONLY: action=list. Approving/rejecting/undoing inbox items lives in stage_automation_review.ConnectorOAuth