PCP

Provides access to PCP metrics through MCP-friendly endpoints, enabling querying of CPU, memory, disk, network, and process data.

python

GitHub Stars

python

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "major-pcp-mcp": {
      "command": "uvx",
      "args": [
        "pcp-mcp"
      ],
      "env": {
        "PCP_HOST": "pmproxy host (default: localhost)",
        "PCP_PORT": "pmproxy port (default: 44322)",
        "PCP_TIMEOUT": "Request timeout in seconds (default: 30)",
        "PCP_USE_TLS": "Use HTTPS for pmproxy (default: false)",
        "PCP_PASSWORD": "HTTP basic auth password (optional)",
        "PCP_USERNAME": "HTTP basic auth user (optional)",
        "PCP_TLS_VERIFY": "Verify TLS certificates (default: true)",
        "PCP_TARGET_HOST": "Target pmcd host to monitor (default: localhost)",
        "PCP_ALLOWED_HOSTS": "Host specifications allowed via host param (optional)",
        "PCP_TLS_CA_BUNDLE": "Path to custom CA bundle (optional)"
      }
    }
  }
}

You can query PCP metrics from your system or remote hosts using an MCP server built to bridge Performance Co-Pilot data into MCP clients. This server enables you to pull CPU, memory, disk I/O, network, processes, and other metrics through a simple, fast interface that MCP clients expect, with secure options and configurable targets.

How to use

You will run the MCP server and connect your MCP client to it. Start with either the fast, no-install option or install the package for persistent use. You can monitor the local machine by default or point the server at a remote PCP instance to collect metrics from another host.

How to install

Prerequisites: you need Python 3.10 or newer and PCP (Performance Co-Pilot) with pmcd and pmproxy running on the target or local system.

# Quick start without installation
uvx pcp-mcp

# Or install as a persistent global tool
uvx tool install pcp-mcp
pcp-mcp

Configuration and usage notes

Configure how you connect to PCP by setting environment variables or using command line options. You can monitor the local host by default or specify a target host to gather metrics from another PCP instance.

Environment variables you may configure include the PCP host, ports, TLS usage, authentication, and timeouts. See examples below for common setups.

Additional notes

This MCP server supports remote monitoring by directing queries to a different pmcd instance via pmproxy. You can enable TLS for secure transport or disable TLS for plain HTTP as needed. Use the provided host and port settings to point at the correct PCP endpoints.

Available tools

get_system_snapshot

Fetches a point-in-time overview of system metrics including CPU, memory, disk, network, and load.

get_process_top

Retrieves the top processes by CPU, memory, or I/O usage.

query_metrics

Fetches current values for specific PCP metrics.

search_metrics

Discovers available metrics by name pattern.

describe_metric

Provides detailed metadata about a specific metric.

Built by

VeilStrat

AI signals for GTM teams