Trigger.dev MCP Server

Trigger.dev runs background tasks and AI agents in TypeScript. Trigger and replay runs, inspect traces and errors, manage schedules, queues and deployments, and query run data.

Tools

Last Updated

Nov 1, 2026

What is the Trigger.dev MCP Server?

The Trigger.dev MCP server gives AI agents structured, permission-aware access to Trigger.dev through the Model Context Protocol. With 47 pre-built actions, agents can read, create, and update Trigger.dev data on behalf of authorized users.

Willow ships the Trigger.dev MCP server as part of an enterprise control plane. Every call runs behind SSO (Okta, Azure AD), enforces RBAC and least-privilege at runtime, writes to a full audit trail, and integrates with Splunk and Loki for SIEM visibility. Connect from Claude Desktop, Claude Code, Cursor, ChatGPT, VS Code, n8n, or any custom agent. Install once, distribute org-wide, and see exactly how Trigger.dev is being used by every AI agent in your stack.

Tools

List Runs

Lists runs in the environment of the API key, newest first. Without a created-at filter only the last 7 days are returned. Use it to browse or report on runs, not to poll a single run.

Get Run

Gets one run with its status, payload, output, tags, metadata, and related runs. Use it to inspect a run you already have the id of.

Get Run Result

Gets the final result of a finished run: ok, output, or error. Returns 404 while the run is still going.

Get Run Trace

Gets the OpenTelemetry trace of a run: its span tree with messages, durations, errors, and events. Use it to find where a run failed or slowed down.

Get Run Events

Lists the OpenTelemetry span events of a run, including log lines. Use it for a flat view of what the run logged.

Cancel Run

Cancels a run that is still in progress. Has no effect on a finished run.

Replay Run

Creates a new run from an existing one, in the same environment. It reuses the original payload and options unless you override them.

Reschedule Run

Changes the delay of a run that is in the DELAYED state. Fails for runs in any other state.

Add Run Tags

Adds tags to a run. A run holds at most 10 tags, and duplicates are ignored.

Update Run Metadata

Replaces the metadata object of a run. Send the full object you want stored.

Trigger Task

Triggers a run of a deployed task in the environment of the API key. Returns the run id.

Batch Trigger Tasks

Triggers many runs at once, each with its own task and payload. Returns the batch id and the run ids.

Get Batch

Gets a batch with its status, run ids, and success and failure counts.

Get Batch Results

Gets the results of the finished runs in a batch. Runs that are still going are left out.

List Schedules

Lists schedules, both declarative and imperative, with their cron expression, timezone, next run, and active flag.

Get Schedule

Gets one schedule by id.

Create Schedule

Creates an imperative cron schedule for a task.

Update Schedule

Updates the task, cron expression, timezone, or external id of an imperative schedule. Declarative schedules defined in code cannot be changed here.

Delete Schedule

Deletes an imperative schedule. Declarative schedules defined in code cannot be deleted here.

Activate Schedule

Turns an imperative schedule back on so it fires again.

Deactivate Schedule

Pauses an imperative schedule so it stops firing. It stays defined and can be activated again.

List Deployments

Lists deployments of the environment, newest first, with version, status, and git info.

Get Deployment

Gets one deployment with its status, version, image, any error, and the tasks in its worker.

Get Latest Deployment

Gets the latest unmanaged deployment of the project: its status, version, and any error.

Promote Deployment

Makes a previously deployed version the current one, so new runs use it. Use it to roll back or roll forward.

List Errors

Lists error groups: failed runs grouped by error type, message, and stack trace, with counts and first and last seen.

Get Error

Gets one error group with its state, counts, and the worker versions it affected.

Resolve Error

Marks an error group as resolved, optionally recording the worker version that fixed it.

Ignore Error

Marks an error group as ignored. Set a duration or thresholds to have it come back when exceeded. With none set, it is ignored indefinitely.

Unresolve Error

Moves a resolved or ignored error group back to unresolved.

List Queues

Lists the queues of the environment with running and queued counts, paused state, and concurrency limits.

Get Queue

Gets one queue by id, or by name together with its type.

Pause or Resume Queue

Pauses a queue so no new runs start, or resumes it. Runs already executing finish.

Override Queue Concurrency

Overrides the concurrency limit of a queue, for example to scale it up or down for a while. Reset it to go back to the limit in code.

Reset Queue Concurrency

Removes a concurrency override so the queue goes back to the limit defined in code.

Execute Query

Runs a TRQL query, a SQL-style language over runs, metrics, and LLM usage, and returns the rows. Call Get Query Schema first to see the tables and columns.

Get Query Schema

Gets the tables and columns TRQL queries can use, with types, descriptions, and allowed values.

List Dashboards

Lists the built-in and custom dashboards with their widgets and TRQL queries, e.g. run success rates, costs, and LLM usage.

Create Waitpoint Token

Creates a waitpoint token. A run waits on it with wait.forToken() until something completes it, either through Complete Waitpoint Token or an HTTP POST to the returned url.

List Waitpoint Tokens

Lists waitpoint tokens of the environment, newest first, with status and output.

Get Waitpoint Token

Gets one waitpoint token with its status and, once completed, its output.

Complete Waitpoint Token

Completes a waitpoint token, which unblocks the run waiting on it. The optional data is returned to that run. Completing a finished token does nothing.

List Environment Variables

Lists the environment variables of a project environment with their values.

Get Environment Variable

Gets the value of one environment variable.

Create Environment Variable

Creates an environment variable in a project environment.

Update Environment Variable

Changes the value of an existing environment variable. Deployed tasks pick up the new value.

Delete Environment Variable

Deletes an environment variable from a project environment.

Customize Tools

Edit descriptions, modify arguments, select tools, or add new ones

Set Up Your Trigger.dev MCP Server in Minutes

Add the following configuration to your MCP client. Authentication is handled via OAuth. Compatible with Claude Desktop, Claude Code, Cursor, ChatGPT, VS Code, n8n, and any MCP-compatible agent.

Claude Desktop

claude_desktop_config.json

{
  "mcpServers": {
    "willow-trigger-dev": {
      "type": "http",
      "url": "https://<org>.mcp-s.com/mcp/mcp/trigger-dev"
    }
  }
}

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "willow-trigger-dev": {
      "type": "http",
      "url": "https://<org>.mcp-s.com/mcp/mcp/trigger-dev"
    }
  }
}

Claude Code

CLI

claude mcp add willow-trigger-dev --transport http https://<org>.mcp-s.com/mcp/mcp/trigger-dev

n8n

HTTP Request Node

{
  "url": "https://<org>.mcp-s.com/mcp/mcp/trigger-dev",
  "method": "POST"
}

Or click "Install with Willow" above to set up automatically with SSO and RBAC preconfigured.

Enterprise Governance for Trigger.dev

Willow adds the layer Trigger.dev and every other SaaS doesn't ship out of the box: every call runs behind SSO (Okta, Azure AD), enforces RBAC and least-privilege at runtime, writes to full audit logs, and detects shadow AI usage across your stack. One MCP gateway. Any agent. Every tool.

Trigger.dev MCP Server FAQ

What is the Trigger.dev MCP server?

The Trigger.dev MCP server is a Model Context Protocol implementation that lets AI agents like Claude, Cursor, and ChatGPT read and write Trigger.dev data through a standardized interface. Willow hosts and governs this server so enterprises can roll it out without a security review backlog.

How is Willow's Trigger.dev MCP server different from the official one?

The official Trigger.dev MCP server is scoped to a single user's account and does not include enterprise governance. Willow's version adds SSO, RBAC, audit logging, shadow AI detection, and centralized control over which actions agents can take across the entire org.

Which AI clients work with the Trigger.dev MCP server?

Claude Desktop, Claude Code, Cursor, ChatGPT, VS Code with MCP support, n8n, and any custom agent built with OpenAI Agents SDK, LangChain, Vercel AI SDK, or Anthropic SDK.

Is the Trigger.dev MCP server secure? How does Willow handle authentication?

Every call runs behind your existing SSO (Okta, Azure AD). Per-user OAuth scopes the agent to exactly what that user can do in Trigger.dev, nothing more. No credentials reach the LLM. Every action writes to an audit trail.

Can I limit which Trigger.dev actions agents can take?

Yes. Willow lets you scope agents to specific actions, specific projects, or specific environments. Toggle actions on or off in the dashboard, or enforce policy via infrastructure-as-code through GitHub.

How do I detect shadow Trigger.dev MCP servers in my org?

Willow's browser extension and discovery service surface unmanaged MCP servers, skills, and AI agents across the org. If a developer installed an unapproved Trigger.dev MCP locally, you'll see it.

What does the Trigger.dev MCP server cost?

Pricing depends on org size and deployment model (SaaS, dedicated cloud, self-host). See withwillow.ai/pricing or contact sales for a quote.

How do I install the Trigger.dev MCP server with Willow?

Install via the Willow Connect Panel in one click, or paste the JSON snippet above into your Claude Desktop, Cursor, or Claude Code config. SSO and RBAC inherit from your existing Willow setup.