prometheus-mcp-server

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

Source: GitHub - pab1it0/prometheus-mcp-server

Features

• Execute PromQL Queries:
  • Execute instant and range PromQL queries against a Prometheus server.
• Metric Discovery and Exploration:
  • List all available metrics in Prometheus.
  • Retrieve metadata for specific metrics.
  • View instant query results.
  • View range query results with different step intervals.
• Authentication Support:
  • Basic authentication via environment variables.
  • Bearer token authentication via environment variables.
• Multi-tenancy Support:
  • Supports Org ID for multi-tenant setups (e.g., Cortex, Mimir, Thanos).
• Docker Support:
  • Official Dockerfile and docker-compose configuration for easy deployment.
  • Multi-stage Docker build for proper MCP communication.
• Configurable Toolset:
  • Tools available to the MCP client are configurable to limit context window usage.
• Comprehensive Test Suite:
  • Includes configuration validation, server functionality, error handling, and main application tests.
• Interactive Tools for AI Assistants:
  • MCP tools include:
    • execute_query: Execute instant PromQL queries.
    • execute_range_query: Execute range queries with time intervals.
    • list_metrics: List all available metrics.
    • get_metric_metadata: Get metadata for a specific metric.
    • get_targets: Get information about all scrape targets.
• Development-Friendly:
  • Uses uv for dependency management.
  • Source structure with clear separation of main application and server logic.
• Open Source:
  • Licensed under the MIT License.

Pricing

No pricing information is provided. The project is open source under the MIT license.