jenkins -enterprise

This MCP server provides enterprise-grade Jenkins interaction through AI-assisted failure analysis, multi-Jenkins routing, and scalable log handling. It is designed to manage many Jenkins instances from a single MCP server, automatically routing analysis requests to the correct Jenkins host based on the full job URL. You’ll be able to diagnose complex build failures, traverse deep sub-build hierarchies, and retrieve targeted log sections from massive pipelines. The server supports advanced AI-driven diagnostics, customizable failure pattern rules, and semantic search to surface relevant context across historical builds. To use it, connect your AI assistant (like Claude) to the MCP server and issue natural-language queries that include the full Jenkins URL, the job path, and optionally a build number. The system will return structured results with root-cause analysis, recommended remediation steps, and relevant log excerpts.

Prerequisites:

• Python 3.10+ installed on your system
• Access to your Jenkins instances (URL, username, API token)
• Optional: Docker and Docker Compose for production deployment

Step-by-step installation:

1. Create and activate a Python virtual environment (optional but recommended):

python3 -m venv venv
source venv/bin/activate  # on macOS/Linux
venv\Scripts\activate     # on Windows

2. Install the MCP package from PyPI (recommended):

pip install jenkins_mcp_enterprise

Optional: enable vector/semantic search (requires additional ML dependencies and a local Qdrant instance):

pip install "jenkins_mcp_enterprise[vector]"

3. Create a configuration directory and a sample config file:

mkdir -p config
cp config/mcp-config.example.yml config/mcp-config.yml

4. Configure your Jenkins instances in config/mcp-config.yml. Example snippet:

jenkins_instances:
  production:
    url: "https://jenkins.yourcompany.com"
    username: "your.email@company.com"
    token: "your-api-token"
    display_name: "Production Jenkins"
settings:
  fallback_instance: "production"

5. Launch the MCP server:

python -m jenkins_mcp_enterprise --config config/mcp-config.yml

6. Optional: connect to Claude Desktop or your AI assistant by configuring the MCP server in your assistant’s config, e.g., adding the server with the appropriate command and config path.

Notes:

• If you prefer a Docker-based deployment, you can adapt the configuration to run the entrypoint with the same config file inside a container.
• Ensure network access between the MCP server and all Jenkins instances, including any required API tokens and SSL settings.

Tips and common considerations:

• Use a centralized config/mcp-config.yml to manage multiple Jenkins instances (production, development, EU, etc.).
• For large logs, enable the vector/semantic search feature only if you have the necessary ML dependencies and a running vector store.
• Ensure API tokens have appropriate scopes and are rotated regularly for security.
• If multi-Jenkins routing isn’t behaving as expected, verify that the full Jenkins job URL matches one of the configured instances and that DNS resolves correctly.
• In production, run the MCP server under a process manager (systemd, pm2, etc.) and enable health checks and logging to monitor uptime.