Get the FREE Ultimate OpenClaw Setup Guide →

trigger.dev

Trigger.dev – build and deploy fully‑managed AI agents and workflows

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio triggerdotdev-trigger.dev docker run -i triggerdev/self-hosted:latest

How to use

Trigger.dev offers a self-hosted platform for building and running AI workflows with durable tasks, retries, queues, and real-time observability. This MCP server description corresponds to the Trigger.dev self-hosted image, which lets you deploy and operate Trigger.dev in your own environment. The platform empowers you to author tasks in TypeScript/JavaScript, connect them to a robust runtime, and manage long-running AI workflows with built-in retries, queues, and observability. You can deploy tasks in your codebase, orchestrate complex AI agent workflows, and monitor runs with full traceability and live updates. The self-hosted instance provides features like human-in-the-loop waits, real-time updates, and multi-environment deployment across DEV, PREVIEW, STAGING, and PROD, all hosted on your infrastructure.

To use it, run the self-hosted server in a container and expose the API and UI so your teams can develop, deploy, and monitor tasks. Connect your codebase by installing the Trigger.dev SDK in your project, define tasks with the task() API, and leverage features like durable cron schedules, batch triggering, and run metadata. The platform is designed to work with your existing toolchains, allowing you to run Python scripts, FFmpeg, browsers, and other system packages inside your tasks, while providing observability, retries, checkpoints, and versioned deployments.

With the self-hosted setup, you’ll gain full control over environments and data, and you can integrate Trigger.dev’s real-time capabilities into your apps and dashboards. Use the UI to inspect runs, configure alerts, and manage resources, or integrate via the SDK in your TypeScript/JavaScript code to define and manage tasks programmatically.

How to install

Prerequisites

  • Docker installed on your host (and optionally Docker Compose for orchestration)
  • Sufficient disk space and memory for the Trigger.dev runtime and stored task data

Step-by-step installation

  1. Pull the self-hosted Trigger.dev image docker pull triggerdev/self-hosted:latest

  2. Run the container (basic example) docker run -d
    --name triggerdev-selfhosted
    -p 8080:8080
    -p 9229:9229
    -e TRIGGERDEV_API_KEY=your-api-key
    -e DATABASE_URL=postgres://user:pass@db:5432/triggerdev
    triggerdev/self-hosted:latest

  3. Optional: Use Docker Compose (recommended for production) cat > docker-compose.yml << 'YAML' version: '3.8' services: triggerdev: image: triggerdev/self-hosted:latest container_name: triggerdev-selfhosted ports: - "8080:8080" - "9229:9229" environment: - TRIGGERDEV_API_KEY=your-api-key - DATABASE_URL=postgres://user:pass@db:5432/triggerdev YAML docker-compose up -d

  4. Initialize and configure

    • Open http://localhost:8080 to access the Trigger.dev UI
    • Create a project, connect your repository, and install the Trigger.dev SDK in your codebase
    • Define and deploy tasks using the SDK (TypeScript/JavaScript) and manage environments, retries, and observability from the UI
  5. Optional persistence and databases

    • Attach a database service (e.g., Postgres) and configure DATABASE_URL in your environment to persist runs, logs, and metadata

Notes

  • Ensure the host ports you expose (default 8080 for UI/API) are available
  • Secure your deployment with proper authentication and network access controls
  • Review the self-hosting docs for environment-specific configuration and upgrade procedures

Additional notes

Tips and common issues:

  • If the UI does not load, verify that port mappings are correct and the container is running. Check logs with docker logs triggerdev-selfhosted.
  • Ensure DATABASE_URL points to a persistent database; running without persistence will lose run history on container recreation.
  • Set TRIGGERDEV_API_KEY to a secure value and limit access to the UI/API.
  • When updating to a new image, perform a controlled upgrade and migrate the database if required.
  • For development, you can run with docker-compose to ease local testing and teardown.
  • If you need to run additional system packages (e.g., browsers, FFmpeg) inside tasks, ensure those tools are available within the container or via configured extensions.

Related MCP Servers

Sponsor this space

Reach thousands of developers