CLIO Node Hardware

io.github.iowarp/node-hardware-mcp

Documentation

Node Hardware MCP Server

Part of CLIO Kit - Gnosis Research Center

Node Hardware MCP is a Model Context Protocol server that enables LLMs to monitor and analyze system hardware information including CPU specifications, memory usage, disk performance, network interfaces, GPU details, and sensor data for both local and remote nodes via SSH connections, providing c...

Quick Start

uvx clio-kit mcp-server node-hardware

Documentation

Full Documentation: CLIO Kit Website
Installation Guide: See INSTALLATION.md
Contributing: See Contribution Guide

Description

🛠️ Installation

Requirements

Python 3.10 or higher
uv package manager (recommended)
SSH client (for remote node capabilities)

Install in Cursor

Go to: Settings -> Cursor Settings -> MCP -> Add new global MCP server

Pasting the following configuration into your Cursor ~/.cursor/mcp.json file is the recommended approach. You may also install in a specific project by creating .cursor/mcp.json in your project folder. See Cursor MCP docs for more info.

{
  "mcpServers": {
    "node-hardware-mcp": {
      "command": "uvx",
      "args": ["clio-kit", "mcp-server", "node-hardware"]
    }
  }
}

Install in VS Code

Add this to your VS Code MCP config file. See VS Code MCP docs for more info.

"mcp": {
  "servers": {
    "node-hardware-mcp": {
      "type": "stdio",
      "command": "uvx",
      "args": ["clio-kit", "mcp-server", "node-hardware"]
    }
  }
}

Install in Claude Code

Run this command. See Claude Code MCP docs for more info.

claude mcp add node-hardware-mcp -- uvx clio-kit mcp-server node-hardware

Install in Claude Desktop

Add this to your Claude Desktop claude_desktop_config.json file. See Claude Desktop MCP docs for more info.

{
  "mcpServers": {
    "node-hardware-mcp": {
      "command": "uvx",
      "args": ["clio-kit", "mcp-server", "node-hardware"]
    }
  }
}

Manual Setup

Linux/macOS:

CLONE_DIR=$(pwd)
git clone https://github.com/iowarp/clio-kit.git
uv --directory=$CLONE_DIR/clio-kit/clio-kit-mcp-servers/node-hardware run node-hardware-mcp --help

Windows CMD:

set CLONE_DIR=%cd%
git clone https://github.com/iowarp/clio-kit.git
uv --directory=%CLONE_DIR%\clio-kit\clio-kit-mcp-servers\node-hardware run node-hardware-mcp --help

Windows PowerShell:

$env:CLONE_DIR=$PWD
git clone https://github.com/iowarp/clio-kit.git
uv --directory=$env:CLONE_DIR\clio-kit\clio-kit-mcp-servers\node-hardware run node-hardware-mcp --help

Capabilities

`get_cpu_info`

Description: Get CPU specifications, core counts, frequencies, and per-core usage. Hints: read-only, idempotent Tags: cpu, hardware

`get_memory_info`

Description: Get RAM and swap capacity, usage percentages, and availability. Hints: read-only, idempotent Tags: hardware, memory

`get_system_info`

Description: Get OS details, hostname, uptime, and active users. Hints: read-only, idempotent Tags: hardware, system

`get_disk_info`

Description: Get disk partitions, usage statistics, and I/O counters. Hints: read-only, idempotent Tags: disk, hardware

`get_network_info`

Description: Get network interfaces, IP addresses, and I/O statistics. Hints: read-only, idempotent Tags: hardware, network

`get_gpu_info`

Description: Get GPU model, memory, temperature, and utilization via nvidia-smi/rocm-smi. Hints: read-only, idempotent Tags: gpu, hardware

`get_sensor_info`

Description: Get temperature, fan speed, and battery sensor readings. Hints: read-only, idempotent Tags: hardware, sensor

`get_process_info`

Description: Get running processes with CPU, memory, and status details. Hints: read-only, idempotent Tags: hardware, process

`get_performance_info`

Description: Get real-time CPU, memory, disk, and network performance metrics. Hints: read-only, idempotent Tags: hardware, performance

`get_remote_node_info`

Description: Collect hardware info from a remote node via SSH. Supports component filtering. Hints: read-only, idempotent Tags: hardware, remote, ssh

`health_check`

Description: Verify server health and hardware monitoring capability status. Hints: read-only, idempotent Tags: diagnostics, hardware

Resources

node-hardware://system-info - Basic system identification info.

Prompts

system_health_check: Guided workflow for a full system health check.

Claude Code

claude mcp add clio-node-hardware -- uvx clio-kit node-hardware

Or install via the CLIO Kit plugin marketplace:

/plugin marketplace add iowarp/clio-kit
/plugin install clio-node-hardware@iowarp-clio-kit

Claude Desktop

Add to your Claude Desktop config (claude_desktop_config.json):

{
  "mcpServers": {
    "clio-node-hardware": {
      "command": "uvx",
      "args": [
        "clio-kit",
        "node-hardware"
      ]
    }
  }
}

Gemini CLI

Add to ~/.gemini/settings.json:

{
  "mcpServers": {
    "clio-node-hardware": {
      "command": "uvx",
      "args": [
        "clio-kit",
        "node-hardware"
      ]
    }
  }
}

Or install the CLIO Kit extension:

gemini extensions install https://github.com/iowarp/clio-kit

Examples

1. Local Hardware Overview

I need a comprehensive overview of my local system's hardware including CPU, memory, disk, and network components.

Tools called:

get_node_info - Get complete local hardware information with all components
Components collected: cpu, memory, disk, network, system, summary

2. Remote Server Monitoring

Monitor the hardware status of a remote server via SSH, focusing on CPU and memory utilization for performance analysis.

Tools called:

get_remote_node_info - Connect to remote host with SSH authentication
Components collected: cpu, memory, performance, system

3. GPU and Thermal Monitoring

Check GPU specifications and thermal sensors on both local and remote systems for machine learning workloads.

Tools called:

get_node_info - Local GPU and sensor monitoring
get_remote_node_info - Remote GPU and thermal analysis
Components collected: gpu, sensors, performance

4. System Health Assessment

Perform a comprehensive health check of system capabilities and verify all monitoring tools are working correctly.

Tools called:

health_check - System health verification and diagnostic assessment
get_node_info - Comprehensive local system analysis with health metrics

5. Performance Bottleneck Analysis

Identify performance bottlenecks on a production server by analyzing CPU, memory, disk I/O, and running processes.

Tools called:

get_remote_node_info - Remote performance analysis via SSH
Components collected: cpu, memory, disk, performance, processes

6. Storage and Network Analysis

Analyze storage health and network interface performance on multiple systems for infrastructure monitoring.

Tools called:

get_node_info - Local storage and network analysis
get_remote_node_info - Remote storage and network monitoring
Components collected: disk, network, system, summary