Prometheus MCP Server

@pab1it0

Basic information

About Prometheus MCP Server

A Model Context Protocol (MCP) server that enables AI agents and LLMs to query and analyze Prometheus metrics through standardized interfaces.

#devops #ai #mcp #prometheus #llm

Prometheus MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "prometheus-mcp-server": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "\\"
      ]
    }
  }
}

Prometheus MCP Server tools

System

Query

Discovery

Overview

What is Prometheus MCP Server?

Prometheus MCP Server is a Model Context Protocol server that gives AI assistants the power to query your Prometheus metrics and execute PromQL queries through standardized MCP interfaces. It is intended for users of MCP-compatible clients such as Claude Desktop, VS Code, Cursor, or Windsurf who need to analyze metrics data from a Prometheus server.

How to use Prometheus MCP Server?

Install and run the server via Docker, or deploy to Kubernetes using the provided Helm chart. Configure it by setting the PROMETHEUS_URL environment variable to point to your Prometheus server, plus optional authentication and transport variables. Add the server to your MCP client’s configuration (for example, Claude Desktop or VS Code) using the Docker command shown in the README.

Key features of Prometheus MCP Server

Execute instant and range PromQL queries against Prometheus
Discover and explore metrics with pagination and filtering
Get metadata for metrics and search by name or description
View scrape target information
Support for basic auth, bearer token, and mutual TLS authentication
Configurable transport modes (stdio, HTTP, SSE)

Use cases of Prometheus MCP Server

AI-assisted monitoring and alert troubleshooting using PromQL
Automated metric exploration and analysis in development workflows
Querying historical metrics for root cause analysis
Integrating Prometheus data with AI-powered dashboards and assistants

Frequently asked questions

What do I need to run Prometheus MCP Server?

You need a Prometheus server accessible from your environment and an MCP-compatible client such as Claude Desktop, VS Code, Cursor, or Windsurf.

How do I connect the server to my Prometheus instance?

Set the `PROMETHEUS_URL` environment variable to the URL of your Prometheus server. Optional authentication can be configured via `PROMETHEUS_USERNAME` and `PROMETHEUS_PASSWORD`, `PROMETHEUS_TOKEN`, or client certificate variables.

What transport protocols does Prometheus MCP Server support?

The server supports stdio (default), HTTP, and SSE transport modes, controlled by the `PROMETHEUS_MCP_SERVER_TRANSPORT` environment variable.

Can I customize which tools are available?

Yes, the list of tools is configurable so you can choose which tools to expose to the MCP client, helping to save context window space.

Is authentication supported?

Yes, the server supports basic authentication, bearer token authentication, and mutual TLS authentication, all configured via environment variables.

Comments

More Developer Tools MCP servers

Vibgrate MCP

Vibgrate

Query your team's drift, vulnerability, and migration data from any AI assistant. Vibgrate MCP connects Cursor, Claude, ChatGPT, Windsurf, or VS Code to Vibgrate Cloud: 51 tools for DriftScores, CVEs and EOL runtimes, up

Nizh

Compliance frameworks delivered to AI agents. Nizh gives your agent read access to your organization's compliance program — SOC 2, ISO 27001, CMMC 2.0, NIST, and more — as MCP tools, so it can check posture, read a contr

GripForge

gripforgeai

The tools your AI needs to make your games. Turn prompts into production-ready game assets — animated characters with their weapons attached, seamless textures, terrain, VFX, HUDs — and playable game kits Unity, Godot, U

Collide MCP

Lithometric Inc. | Collide Inc.

Collide is the multiplayer layer for AI coding agents, and it pays for itself. Expert orchestration across every agent on your repo saves teams tens to hundreds of dollars a day in token burn. Run Claude Code, Codex, Cu

Yocoolab

MCP server for AI Agents — exposes Yocoolab feedback threads and design selections as tools