Caesaryangs Prometheus MCP Server

🚀 MCP Server for Prometheus

An MCP (Model Context Protocol) server for retrieving data from a Prometheus database. This MCP server enables Large Language Models (LLMs) to call executable utility functions in predefined routes, thereby retrieving and analyzing large amounts of metric data, searching for metric usage, performing complex queries, and other related tasks, while enhancing control over usage.

✨ Features

• Data Retrieval: Obtain data for specific metrics or within a specific time period from Prometheus.
• Metric Analysis: Conduct statistical analysis on the retrieved metrics.
• Usage Search: Find and explore metric usage patterns.
• Complex Queries: Execute advanced PromQL queries for in - depth data exploration.

Specific Functionality

✅ Retrieve comprehensive metric information, including names and descriptions, from Prometheus. ✅ Fetch and analyze specific metric data using metric names. ✅ Analyze metric data within a custom time period. 🚧 Filter and match data based on specific labels (Under development). ⏳ Other planned features...

🚀 Quick Start

Running MCP requires a Python virtual environment (venv), and all packages should be installed in this venv so that the MCP server can start automatically.

Prepare the Python Environment

cd ./src/prometheus_mcp_server
python3 -m venv .venv

# Linux/macOS:
source .venv/bin/activate

# Windows:
.venv\Scripts\activate

Then, prepare the Python environment for specific use.

Install Required Packages

Ensure that pip is installed correctly. If your venv does not come with pip installed, install it manually:

wget https://bootstrap.pypa.io/get-pip.py
python3 get-pip.py

Then install all the necessary packages:

pip install -r requirements.txt

💻 Usage Examples

With MCP Client (Including Claude Desktop)

Configure your Claude Desktop application configuration file ~/Library/Application Support/Claude/claude_desktop_config.json (macOS)

{
    "mcpServers": {
        "prometheus": {
            "command": "uv",
            "args": [
                "--directory",
                "/path/to/prometheus_mcp_server",
                "run",
                "server.py"
            ],
            "env": {
                "PROMETHEUS_HOST": "http://localhost:9090"
            }
        }
    }
}

Run the MCP Server Independently

Using the uv Method

uv --directory /path/to/prometheus_mcp_server run server.py

This is also a way to ensure that the MCP server can start automatically, as Claude Desktop uses this ux script method to start the server when the application launches.

Using the Regular Python Method

python3 server.py

🔍 Contribute to the Project

Contributions are welcome! Here are some quick guidelines:

1. Fork the repository.
2. Create a feature branch (git checkout -b feature/AmazingFeature).
3. Commit your changes (git commit -m 'Add some AmazingFeature').
4. Push to the branch (git push origin feature/AmazingFeature).
5. Open a pull request.

For major changes, please open an issue first to discuss what you would like to change.

Thank you for your contributions!

📄 License

MIT License

📚 References and Acknowledgments

This project is inspired by or uses code from the following open - source projects:

• Prometheus API Client - The Prometheus API call code is modified based on this library.
• MySQL MCP Server - A similar database - oriented MCP server implementation.