MCP System Monitor Server

Provides real-time cross-platform system monitoring data through MCP tools, including CPU/GPU/memory/disk/network metrics and phase-1 performance data.

python

0

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": {
    "huhabla-mcp-system-monitor": {
      "command": "uv",
      "args": [
        "run",
        "mcp",
        "dev",
        "mcp_system_monitor_server.py"
      ],
      "env": {
        "YOUR_LONG_LIVED_TOKEN": "ABC123XYZ"
      }
    }
  }
}

You can run and connect to the MCP System Monitor Server to monitor real-time system metrics across CPU, GPU, memory, disk, network, and processes, with optional phase 1 performance data. It exposes a set of MCP tools and resources that let you query live data or run lightweight monitoring sessions from MCP clients such as Claude Desktop. Use the provided start commands to run the server locally or install it into Claude Desktop for quick access.

How to use

You interact with the MCP System Monitor Server through an MCP client. After you start the server locally, you can request current system metrics, trigger short monitoring sessions, or retrieve complete snapshots of system state. Use natural language prompts or predefined tool calls to access data such as current CPU usage, GPU details, memory usage, disk space, network statistics, and top processes by CPU or memory.

How to install

# Prerequisites: ensure you have Python 3.10+ and a supported OS (Windows, macOS, or Linux).
# Optional: you may want to install uv for convenient command line execution.

# Development/Usage notes (choose one flow):
# 1) Run in development mode with the MCP Inspector via uv
uv run mcp dev mcp_system_monitor_server.py

# 2) Install into Claude Desktop for quick access
uv run mcp install mcp_system_monitor_server.py --name "System Monitor"

# 3) Run the server directly from Python
python mcp_system_monitor_server.py

Additional configuration and usage notes

The server also includes example MCP JSON configurations for integrating with MCP clients on Windows and macOS. These snippets show how to register the MCP system monitor in a client by providing a command and arguments that start the local server. Use the exact structures shown in the examples to create your own client configuration.

Example MCP JSON configurations

{
  "mcpServers": {
    "mpc-system-monitor": {
      "command": "cmd",
      "args": [
        "/c",
        "C:/Users/Sören Gebbert/Documents/GitHub/mcp-system-monitor/start_mpc_system_monitor.bat"
      ]
    }
  }
}

{
  "mcpServers": {
    "mpc-system-monitor": {
      "command": "/bin/zsh",
      "args": [
        "/Users/holistech/Documents/GitHub/mcp-system-monitor/start_mcp_system_monitor.sh"
      ]
    }
  }
}

Available tools

get_current_datetime

Return the current local datetime in ISO format.

get_cpu_info

Return current CPU usage and statistics including per-core data.

get_gpu_info

Return GPU information for all detected GPUs.

get_memory_info

Return RAM and swap usage statistics.

get_disk_info

Return disk usage for all mounted drives.

get_system_snapshot

Return a complete snapshot of the system state.

monitor_cpu_usage

Monitor CPU usage over a specified duration.

get_top_processes

Return top processes by CPU or memory usage.

get_network_stats

Return network interface statistics.

get_io_performance

Return detailed I/O performance metrics and rates.

get_system_load

Return system load averages and process statistics.

get_enhanced_memory_info

Return detailed memory statistics including caches and buffers.

get_enhanced_network_stats

Return enhanced network performance metrics.

get_performance_snapshot

Return a complete performance monitoring snapshot.

monitor_io_performance

Monitor I/O performance over a duration with trend analysis.