MCP Jenkins Server

What is MCP Jenkins Server?

MCP Jenkins Server is a Model Context Protocol server that integrates with Jenkins, providing tools to interact with Jenkins jobs, builds, views, and server information. It is intended for developers using MCP-compatible clients who need programmatic access to Jenkins automation infrastructure.

How to use MCP Jenkins Server?

Configuration requires a .env file with JENKINS_URL, JENKINS_USER, and JENKINS_TOKEN (or alternate env variables JENKINS_USERNAME and JENKINS_PASSWORD in CLI setup). The server can be launched via uv with the provided JSON MCP configuration. Python 3.10+ and a Jenkins server with API access are required. Example client code: from mcp.client import Client; client = Client("http://localhost:8000"); response = client.call_tool("list_jobs").

Key features of MCP Jenkins Server

• Retrieve Jenkins server information.
• List and inspect Jenkins jobs.
• Get build details and console output.
• Manage and list Jenkins views.
• Trigger specific job builds.

Use cases of MCP Jenkins Server

• Automating CI/CD status checks from an AI assistant.
• Fetching build logs for debugging without opening the Jenkins UI.
• Triggering a demo job (e.g., “LLM_Demo”) with user parameters.
• Listing all jobs in a Jenkins view for project overview.
• Monitoring build history for a specific job.

FAQ from MCP Jenkins Server

What is MCP Jenkins Server and what does it do?

It is an MCP server that exposes Jenkins functions as tools, allowing MCP clients (such as AI assistants) to query Jenkins server info, list jobs, inspect builds, manage views, and trigger builds.

How do I configure MCP Jenkins Server?

Create a .env file with JENKINS_URL, JENKINS_USER, and JENKINS_TOKEN. Alternatively, set environment variables JENKINS_URL, JENKINS_USERNAME, and JENKINS_PASSWORD in the MCP client configuration.

What are the system requirements?

Python 3.10 or later, a Jenkins server with API access enabled, and a valid Jenkins API token (or username/password) for authentication.

What tools are available?

The server provides seven tools: get_jenkins_info, list_jobs, get_job_info (parameter: job_name), get_build_info (parameters: job_name, build_number), get_build_console_output (parameters: job_name, build_number), get_views, and trriger_llm_demo_job_build (parameter: user).

Which transport or client setup is supported?

The example client uses an HTTP endpoint (http://localhost:8000). The CLI configuration launches the server using uv with the server.py script, which uses stdio MCP transport. Both approaches are described in the README.