pmmcp

performancecopilot/pmmcp
★ 0 stars Python AI/LLM Updated today
MCP server for Performance Co-Pilot(PCP) - allow performance analysis with natural language via Claude Code
View on GitHub → Try with Claude — $10 free →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "pmmcp": {
      "command": "uvx",
      "args": [
        "pmmcp"
      ]
    }
  }
}

Or install with pip: pip install pmmcp

README Excerpt

<p align="center"> <img src="logo.jpg" width="700"> </p> An [MCP](https://modelcontextprotocol.io) server that exposes [Performance Co-Pilot (PCP)](https://pcp.io) metrics to AI agents via [pmproxy](https://man7.org/linux/man-pages/man1/pmproxy.1.html). Enables Claude and other LLMs to query live and historical performance data from any PCP-monitored host.

Tools (20)

PMMCP_GRAFANA_FOLDERPMMCP_HOSTPMMCP_PORTPMMCP_REPORT_DIRPMMCP_TRANSPORTPMPROXY_TIMEOUTPMPROXY_URLcompare_periodscoordinate_investigationfleet_health_checkincident_triageinvestigate_subsystempcp_compare_windowspcp_derive_metricpcp_discover_metricspcp_fetch_livepcp_fetch_timeseriespcp_get_hostspcp_get_metric_infopcp_query_series