Prometheus MCP Server

Add to your client

Copy the config for your MCP client and paste it into its config file.

{
  "mcpServers": {
    "prometheus-mcp-server": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "PROMETHEUS_URL",
        "ghcr.io/pab1it0/prometheus-mcp-server:latest"
      ],
      "env": {
        "PROMETHEUS_URL": "<your-prometheus-url>"
      }
    }
  }
}

Before you start

• A reachable Prometheus server URL, set via the PROMETHEUS_URL environment variable
• Docker (recommended) — or Python 3.10+ with the uv package manager for a local install
• An MCP-compatible client (Claude Desktop, VS Code, Cursor, Windsurf, etc.)
• Optional credentials only if your Prometheus is secured: basic auth (PROMETHEUS_USERNAME/PASSWORD), bearer token (PROMETHEUS_TOKEN), mTLS client cert/key, or custom headers

About Prometheus MCP Server

Tools & capabilities (6)

When to use it

• Use it when you want to ask for a metric in plain English and have the agent write and run the PromQL
• Use it when investigating an alert and you need a range query plotted over the last hour
• Use it when exploring an unfamiliar Prometheus — list metrics and inspect metadata to discover what's available
• Use it when checking whether scrape targets are up before debugging missing data
• Use it when you run multiple Prometheus environments and want them namespaced via TOOL_PREFIX in one client

Quick setup

1. 1Pull the Docker image (or install the Python package with uv)
2. 2Set PROMETHEUS_URL to your Prometheus endpoint, plus any auth env vars if it's secured
3. 3Add the server to your MCP client config (stdio by default, or HTTP/SSE if running it as a service)
4. 4Restart the client so it loads the server and discovers the tools
5. 5Run health_check or list_metrics to confirm connectivity

Prometheus MCP Server FAQ