Prometheus MCP Server

🚀 Prometheus MCP Server

A Model Context Protocol (MCP) server that fetches data from the Prometheus database. This MCP server enables Large Language Models (LLMs) to call utility functions for retrieving and analyzing large amounts of metric data, searching for metric usage, performing complex queries, and other related tasks.

🚀 Quick Start

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

Installation via Smithery

To automatically install the Prometheus MCP Server for Claude Desktop via Smithery:

npx -y @smithery/cli install @CaesarYangs/prometheus_mcp_server --client claude

Manual Installation

Prepare the Python Environment

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

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

# Windows:
.venv\Scripts\activate

Then, you're ready to use the dedicated Python environment.

Install the Required Packages

Ensure that pip is installed correctly. If your venv was installed without pip, install it manually:

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

Then install all the required packages:

pip install -r requirements.txt

✨ Features

• Data Retrieval: Fetch specific metrics from Prometheus by name or time range.
• Metric Analysis: Perform 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.

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

💻 Usage Examples

With Cursor Env

Get ready for updates according to the more user - friendly Cursor environment.

Set the location of this MCP server in the Cursor settings here:

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

With MCP Client (Including Claude Desktop)

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

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

Stand - alone MCP Server

Start this 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 runs correctly.

Run Independently

python3 src/prometheus_mcp_server/server.py

🤝 Contribution

If you'd like to contribute to the project, please visit here to submit your PR or Issue.

📄 License

[License information to be filled in here]

📚 Documentation

• Prometheus Official Documentation
• MCP Protocol