prometheus-mcp and prometheus_mcp_server

These two tools are competitors, as both implement a Model Context Protocol (MCP) server designed to provide AI agents or LLMs with programmatic access to Prometheus metrics.

prometheus-mcp
56
Established
prometheus_mcp_server
40
Emerging
Maintenance 10/25
Adoption 6/25
Maturity 24/25
Community 16/25
Maintenance 0/25
Adoption 7/25
Maturity 16/25
Community 17/25
Stars: 24
Forks: 6
Downloads:
Commits (30d): 0
Language: TypeScript
License: MIT
Stars: 34
Forks: 11
Downloads:
Commits (30d): 0
Language: Python
License: MIT
No risk flags
Stale 6m No Package No Dependents

About prometheus-mcp

idanfishman/prometheus-mcp

A Model Context Protocol (MCP) server implementation that provides AI agents with programmatic access to Prometheus metrics via a unified interface.

This project integrates your AI assistant with Prometheus, allowing you to ask natural language questions about your monitoring data. You can input requests like "What's the CPU usage on server X?" and receive real-time metrics and insights directly from Prometheus. Site reliability engineers, DevOps professionals, and system administrators can use this to quickly understand system performance and health.

site-reliability-engineering devops system-monitoring observability operations

About prometheus_mcp_server

CaesarYangs/prometheus_mcp_server

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

This tool helps Site Reliability Engineers and DevOps professionals understand their system's health and performance by connecting Large Language Models (LLMs) to Prometheus databases. It allows LLMs to retrieve specific metrics, analyze data within custom time ranges, and explore usage patterns without needing to write complex PromQL queries manually. The output includes structured data and analyses from your Prometheus metrics.

Site Reliability Engineering DevOps System Monitoring Incident Response Performance Analysis

Scores updated daily from GitHub, PyPI, and npm data. How scores work