The Trigger.dev MCP (Model Context Protocol) Server gives AI coding assistants direct access to your Trigger.dev projects. Once connected, your assistant can search docs, trigger task runs, deploy projects, and inspect run details — all without leaving the chat.
What the MCP Server can do
Search the Trigger.dev documentation
Initialize Trigger.dev in a new project
List projects, organizations, and tasks
Trigger task runs with custom payloads
Deploy projects to dev, staging, or production
Fetch run details and list runs with status filters
Installation
The fastest way to get started is the interactive installer, which detects your installed AI clients and configures them automatically:
npx trigger.dev@latest install-mcp
You can also pass --yolo to install into all detected clients at once:
npx trigger.dev@latest install-mcp --yolo
Or target specific clients:
npx trigger.dev@latest install-mcp --client claude-code cursor
After installation, restart your AI client. You should see a server named trigger connect automatically.
Client configuration
If you prefer to configure manually, copy the snippet for your client below.
Claude Code
Cursor
Windsurf
VS Code
Zed
Cline
Gemini CLI
opencode
Codex CLI
npx trigger.dev@latest install-mcp --client claude-code
Or add to ~/.claude.json (user) or .mcp.json (project): {
"mcpServers" : {
"trigger" : {
"command" : "npx" ,
"args" : [ "trigger.dev@latest" , "mcp" ]
}
}
}
npx trigger.dev@latest install-mcp --client cursor
Or add to ~/.cursor/mcp.json (user) or .cursor/mcp.json (project): {
"mcpServers" : {
"trigger" : {
"command" : "npx" ,
"args" : [ "trigger.dev@latest" , "mcp" ]
}
}
}
npx trigger.dev@latest install-mcp --client windsurf
Or add to ~/.codeium/windsurf/mcp_config.json: {
"mcpServers" : {
"trigger" : {
"command" : "npx" ,
"args" : [ "trigger.dev@latest" , "mcp" ]
}
}
}
npx trigger.dev@latest install-mcp --client vscode
Or add to .vscode/mcp.json (project) or ~/Library/Application Support/Code/User/mcp.json (user, macOS): {
"servers" : {
"trigger" : {
"command" : "npx" ,
"args" : [ "trigger.dev@latest" , "mcp" ]
}
}
}
VS Code uses servers instead of mcpServers.
npx trigger.dev@latest install-mcp --client zed
Or add to ~/.config/zed/settings.json: {
"context_servers" : {
"trigger" : {
"source" : "custom" ,
"command" : "npx" ,
"args" : [ "trigger.dev@latest" , "mcp" ]
}
}
}
npx trigger.dev@latest install-mcp --client cline
Or add to ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json: {
"mcpServers" : {
"trigger" : {
"command" : "npx" ,
"args" : [ "trigger.dev@latest" , "mcp" ]
}
}
}
npx trigger.dev@latest install-mcp --client gemini-cli
Or add to ~/.gemini/settings.json (user) or .gemini/settings.json (project): {
"mcpServers" : {
"trigger" : {
"command" : "npx" ,
"args" : [ "trigger.dev@latest" , "mcp" ]
}
}
}
npx trigger.dev@latest install-mcp --client opencode
Or add to ~/.config/opencode/opencode.json (user) or ./opencode.json (project): {
"mcp" : {
"trigger" : {
"type" : "local" ,
"command" : [ "npx" , "trigger.dev@latest" , "mcp" ],
"enabled" : true
}
}
}
npx trigger.dev@latest install-mcp --client openai-codex
Or add to ~/.codex/config.toml: [ mcp_servers . trigger ]
command = "npx"
args = [ "trigger.dev@latest" , "mcp" ]
Authentication
The search_docs tool works without any authentication. All other tools require you to be logged in via the Trigger.dev CLI:
npx trigger.dev@latest login
The first time you use an authenticated tool, your AI client will prompt you to complete the login flow in your browser.
Getting started prompts
Once the server is connected, try these prompts in your AI assistant:
"Search the Trigger.dev docs for an ffmpeg example"
"Initialize Trigger.dev in my project"
"List all tasks in my project"
"Trigger my send-email task with a sample payload"
"Get the details of the latest run for my process-video task"
"Deploy my project to staging"
Advanced options
The install-mcp command supports several options for fine-grained control:
Flag Description --client <clients...>Install into specific client(s) --scope user | project | localWhere to write the config --project-ref <ref>Scope the server to a specific project --dev-onlyRestrict to the dev environment only --yoloInstall into all supported clients automatically --tag <tag>Pin to a specific CLI version (default: latest)
Example: restrict to dev environment for a specific project:
npx trigger.dev@latest install-mcp --dev-only --project-ref proj_abc123
To pass options through a manual config, add them to the args array:
{
"args" : [ "trigger.dev@latest" , "mcp" , "--dev-only" , "--project-ref" , "proj_abc123" ]
}
Troubleshooting
If the server shows as disconnected, try restarting your AI client. MCP server connections are
established on startup.
Run npx trigger.dev@latest install-mcp again — it detects and fixes common configuration issues.
Verify the server starts without errors by running npx trigger.dev@latest mcp in your terminal.
Make sure you have restarted the AI client after editing its config file.
Check that node and npx are on your PATH (some clients use a restricted environment).
Next steps
Building with AI Tools and patterns for writing Trigger.dev code with AI assistants.
Quickstart Get started with Trigger.dev and run your first task in minutes.