Skip to main content

Rundeck / Runbook Automation MCP Server

Rundeck / Runbook Automation MCP Server

Ask your AI assistant to "create a job that restarts the web service on all production nodes, and import it," and it generates a schema-valid job definition and submits it to your real Rundeck instance — no context switching to the UI, no hand-written API call, no YAML you have to get right from memory. That's what the Rundeck MCP Server does: it connects any MCP-compatible AI assistant (Claude Desktop, Claude Code, Cursor, VS Code, and others) directly to your Rundeck or Runbook Automation (RBA) instance, authenticated with your own API token.

Why use it

Normally, getting an AI assistant to help with Rundeck means it's guessing: it doesn't know your job schema, your node filter syntax, or your ACL policy format, so you get plausible-looking YAML that fails on import, or advice that's slightly wrong for your version. Two things fix that:

  • It acts, instead of describing. Ask it to list running executions, restart a job, or provision a runner, and it calls the real API and shows you the real result — not a suggested curl command for you to run yourself.
  • It knows the rules before it writes anything. Job schemas, node filter syntax, and ACL policy structure are loaded as context the assistant checks its own output against, so a generated job definition is validated before it ever reaches your instance.

What it can do

  • Answer Rundeck questions on the spot, pulled from the official documentation: API usage, job schemas, node filters, plugin configuration, and more.
  • Query and drive your real instance: look up projects, jobs, executions, and nodes, or trigger a job run, without leaving the chat.
  • Generate and validate job definitions: describe a job in plain language and get back a ready-to-import YAML or JSON definition, checked against Rundeck's schema before you deploy it.
  • Validate and manage ACL policies: catch a structural mistake before it's ever submitted, or list, read, create, update, and delete stored policies directly.
  • Provision runners: create system- or project-scoped Rundeck Runners on demand, on Docker, Kubernetes, Linux, or Windows.

Under the hood, this runs on three MCP building blocks — tools (actions the assistant can call), resources (read-only documentation, addressed via rundeck:// URIs), and prompts (guided, multi-step workflows) — but you shouldn't need to think about that distinction to use it. It only matters if you're building against the protocol directly; see Technical Capabilities for that.

Quick start

The fastest path is the published Docker image:

docker run -i --rm \
  -e RUNDECK_URL=https://your-rundeck-instance.example.com \
  -e RUNDECK_TOKEN=your-rundeck-api-token \
  rundeck/mcp:latest

Wire that command into your MCP client's configuration and you're connected. See Installation for the Claude Code walkthrough, or Other Setups for Claude Desktop, Cursor, VS Code, and the npm-based alternative if you'd rather not run Docker.

In this section

Privacy Policy |