Prometheus MCP Server

Example Use Cases

• 1

  Run instant or range PromQL queries against a Prometheus instance to retrieve metric values and time series.

• 2

  Discover and inspect series, label names and values, and metric metadata to understand metric topology and semantics.

• 3

  Surface operational information such as scrape targets, alerts, recording/alerting rules, and Prometheus status to users or automated workflows.

Description

This project implements a Model Context Protocol (MCP) server in TypeScript that acts as a bridge between Claude (or other MCP-capable agents) and a Prometheus instance. It supports instant and range PromQL queries, series discovery, label exploration, metric metadata, scrape target info, alerts, rules, and status endpoints. Configuration is simple via the PROMETHEUS_BASE_URL environment variable and it can be run locally or via npx. The server is intended to make Prometheus data accessible to agents over stdio/MCP.

Capabilities(5 total)