MCP Server for Prometheus

Name: MCP Server for Prometheus
Author: CaesarYangs

@CaesarYangs

关于 MCP Server for Prometheus

A Model Context Protocol (MCP) server enabling LLMs to query, analyze, and interact with Prometheus databases through predefined routes.

#prometheus #model-context-protocol

基本信息

分类

开发工具

许可证

MIT

运行时

python

传输方式

stdio

发布者

CaesarYangs

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "prometheus_mcp_server": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "@CaesarYangs/prometheus_mcp_server",
        "--client",
        "claude"
      ]
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is MCP Server for Prometheus?

MCP Server for Prometheus is a Model Context Protocol (MCP) server that enables Large Language Models (LLMs) to retrieve and analyze metric data from Prometheus databases. It provides tool functions for fetching specific metrics, performing statistical analysis, searching metric usage, and executing PromQL queries, with predefined routes for controlled usage.

How to use MCP Server for Prometheus?

Install automatically via Smithery using npx -y @smithery/cli install @CaesarYangs/prometheus_mcp_server --client claude, or manually by creating a Python virtual environment, installing dependencies from requirements.txt, and running server.py with uv or Python. Configure the Prometheus host via the PROMETHEUS_HOST environment variable and integrate with MCP clients like Claude Desktop or Cursor by specifying the command and directory in the client's MCP settings.

Key features of MCP Server for Prometheus

Retrieve comprehensive metric information (names, descriptions) from Prometheus.
Fetch and analyze specific metric data by name.
Analyze metric data within custom time ranges.
Execute advanced PromQL queries for in‑depth data exploration.
Search and explore metric usage patterns.

Use cases of MCP Server for Prometheus

Let an LLM answer questions about available metrics and their descriptions.
Analyze time‑series metric data over a custom time window.
Run complex PromQL queries to investigate infrastructure or application performance.
Quickly find metric usage patterns without manual PromQL writing.

FAQ from MCP Server for Prometheus

What is MCP Server for Prometheus and what problem does it solve?

It is an MCP server that allows LLMs to interact with Prometheus databases through tool functions. This enables AI‑assisted retrieval, analysis, and querying of metric data without writing manual PromQL or using the Prometheus UI.

What are the prerequisites to run MCP Server for Prometheus?

You need a Python virtual environment (venv) with pip installed, the dependencies listed in requirements.txt, and access to a running Prometheus instance. The server can be started using uv or standard Python.

How do I configure the Prometheus host?

Set the environment variable PROMETHEUS_HOST to your Prometheus server URL (e.g., http://localhost:9090). This is specified in the MCP client configuration (e.g., Claude Desktop's claude_desktop_config.json) under the env key.

What capabilities are currently available?

The server supports retrieving metric names and descriptions, fetching and analyzing specific metric data, analyzing data within custom time ranges, and executing PromQL queries. Filtering by labels is still in development.

Are there any known limitations or features in development?

Label‑based filtering is not yet implemented (in development). Additional features are planned but not specified. The server relies on an external Prometheus database and requires that database to be reachable.

开发工具分类下的更多 MCP 服务器

Golf

golf-mcp

Production-Ready MCP Server Framework • Build, deploy & scale secure AI agent infrastructure • Includes Auth, Observability, Debugger, Telemetry & Runtime • Run real-world MCPs powering AI Agents

834