pab1it0/prometheus-mcp-server

Overview

An MCP server that lets AI agents and LLMs query Prometheus metrics through a standardized interface. It exposes PromQL queries as tools, enabling natural language interaction with monitoring data. Built in Python, it bridges the gap between LLM workflows and observability infrastructure.

Best for

Best for
Developers building AI agents that need real-time access to Prometheus metrics

Use cases

• Query Prometheus metrics from AI agents or chat interfaces
• Automate incident response by feeding live metrics into LLM analysis
• Integrate monitoring data into natural language dashboards or reports

How to use

Install

uv pip install -e .

Tools exposed

• PROMETHEUS_URL
• PROMETHEUS_URL_SSL_VERIFY
• PROMETHEUS_DISABLE_LINKS
• PROMETHEUS_REQUEST_TIMEOUT
• PROMETHEUS_USERNAME
• PROMETHEUS_PASSWORD
• PROMETHEUS_TOKEN
• PROMETHEUS_CLIENT_CERT
• PROMETHEUS_CLIENT_KEY
• REQUESTS_CA_BUNDLE
• ORG_ID
• PROMETHEUS_MCP_SERVER_TRANSPORT
• PROMETHEUS_MCP_BIND_HOST
• PROMETHEUS_MCP_BIND_PORT
• PROMETHEUS_MCP_STATELESS_HTTP
• PROMETHEUS_CUSTOM_HEADERS
• TOOL_PREFIX
• health_check
• execute_query
• execute_range_query

Tested with

Claude Desktop, Claude Code, Cursor, Windsurf, VS Code

Example client config

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

Notes

An MCP server that lets AI agents and LLMs query Prometheus metrics through a standardized interface. It exposes PromQL queries as tools, enabling natural language interaction with monitoring data. Built in Python, it bridges the gap between LLM workflows and observability infrastructure.

450 stars on GitHub. Last updated 2026-05-19. Licensed MIT.

Use cases

• Query Prometheus metrics from AI agents or chat interfaces
• Automate incident response by feeding live metrics into LLM analysis
• Integrate monitoring data into natural language dashboards or reports

Pros

• Standard MCP protocol makes integration with existing AI frameworks straightforward
• Leverages PromQL for powerful, flexible metric queries
• Active open-source project with 450 stars and community contributions

Cons

• Requires a running Prometheus instance and network access to it
• Limited to read-only queries; cannot modify alert rules or configuration
• MCP ecosystem is still evolving, so client support may vary

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.