PC Manager

AI-powered PC management system with 80+ MCP tools for Cursor, Claude, and other AI agents.

PC Manager bridges the gap between AI and your operating system. It provides a comprehensive REST API, a web dashboard, and an MCP (Model Context Protocol) server that gives AI agents full control over your Windows PC.

Features

System Monitoring

• Real-time CPU, memory, disk, and network monitoring

• Live charts with usage history

• System information and uptime tracking

File Management

• Browse, create, rename, move, copy, delete files and folders

• Built-in file viewer and editor

• Drive listing with usage info

Process & Service Control

• List, filter, and kill processes

• Start, stop, restart Windows services

• Process memory and CPU tracking

Network

• Interface listing with IP addresses

• Active connection monitoring

• Traffic statistics

• Ping, DNS resolve, port scan, Wake-on-LAN

Shell

• Execute PowerShell commands remotely via web UI or API

• Command history with arrow key navigation

Power Management

• Lock, sleep, restart, shutdown with confirmation

• Power plan and battery status

Programs & Environment

• List installed programs and startup items

• View and edit environment variables

• PATH analysis with existence checks

Task Scheduler

• List and run scheduled tasks

Registry

• Read and write Windows registry values

Security & Admin

• API key authentication

• Danger level system (LOW / MEDIUM / HIGH / CRITICAL)

• Action logging and audit history

• Admin Bridge for elevated privilege execution

Hardware & Sensors

• GPU info (NVIDIA)

• Disk health (SMART)

• USB devices, Bluetooth, printers

• Camera capture, microphone recording

Screen & GUI Control

• Screenshot capture

• Mouse click and move

• Keyboard input

• Window management (list, focus)

• Clipboard read/write

Docker & Package Management

• Docker container management

• Winget package search and install

• WiFi network listing and connection

Email Integration

• IMAP/SMTP email access (Naver, Gmail, etc.)

• Read, search, send, delete emails

• Folder listing

AI Agent

• Natural language to API translation

• OpenAI Function Calling compatible schema

• Chat interface in web dashboard

Related MCP server: Conduit

Architecture

+------------------+     +------------------+     +------------------+
|   Web Dashboard  |     |    AI Agents     |     |   External Apps  |
|  (localhost:7777)|     | (Cursor/Claude)  |     |   (GPT/Gemini)   |
+--------+---------+     +--------+---------+     +--------+---------+
         |                         |                        |
         |    REST API (/api/*)    |     MCP Protocol       |
         +------------+------------+----------+-------------+
                      |                       |
              +-------v-------+       +-------v-------+
              |   Flask App   |       |  MCP Server   |
              |   (app.py)    |       |(mcp_server.py)|
              +-------+-------+       +-------+-------+
                      |                       |
              +-------v-----------------------v-------+
              |          Core Modules                  |
              |  system | files | processes | network  |
              |  shell | power | registry | docker    |
              |  screen | senses | email | admin      |
              +---------------------------------------+
              |          Windows OS (PowerShell)       |
              +---------------------------------------+

Quick Start

Prerequisites

• Windows 10/11

• Python 3.11+

• pip

Installation

# Clone the repository
git clone https://github.com/yourusername/pc-manager.git
cd pc-manager

# Install dependencies
pip install -r requirements.txt

# Run the web dashboard
python app.py

Open http://localhost:7777 in your browser.

MCP Server Setup (Cursor / Claude Desktop)

1. Add to your MCP configuration:

{
  "mcpServers": {
    "pc-manager": {
      "command": "python",
      "args": ["path/to/pc-manager/mcp_server.py"]
    }
  }
}

2. Restart Cursor or Claude Desktop

3. You now have 80+ tools available to your AI agent

Email Setup (Optional)

1. Copy the example config:

cp mail_config.example.json mail_config.json

2. Edit mail_config.json with your email credentials:

   • For Naver Mail: Enable IMAP in settings, generate an app password

   • For Gmail: Enable IMAP, use app-specific password

Admin Bridge Setup (Optional)

For operations requiring elevated privileges:

powershell -ExecutionPolicy Bypass -File admin_bridge/setup_admin_bridge.ps1

API Documentation

Once the server is running, visit:

• Swagger UI: http://localhost:7777/api/docs

• OpenAI Schema: http://localhost:7777/api/agent/openai-schema

API Endpoints Overview

MCP Tools (80+)

When connected via MCP, AI agents can use tools including:

Project Structure

pc-manager/
├── app.py                 # Flask entry point
├── config.py              # Configuration
├── mcp_server.py          # MCP server (80+ tools)
├── requirements.txt       # Python dependencies
├── api/                   # REST API modules
│   ├── system.py          # System info & stats
│   ├── files.py           # File management
│   ├── processes.py       # Process management
│   ├── services.py        # Service control
│   ├── network.py         # Network monitoring
│   ├── disk.py            # Disk analysis
│   ├── shell.py           # Shell execution
│   ├── power.py           # Power management
│   ├── agent.py           # AI agent bridge
│   └── ...                # 15+ more modules
├── core/                  # Core utilities
│   ├── security.py        # Authentication
│   ├── history.py         # Action logging
│   └── admin_bridge.py    # Elevated execution
├── admin_bridge/          # Admin elevation scripts
├── static/                # Frontend assets
│   ├── css/style.css
│   └── js/app.js
├── templates/
│   └── index.html         # Web dashboard
└── .cursor/
    └── mcp.json           # MCP configuration

Security Notice

• The API runs on localhost only by default

• All operations are authenticated via API key

• Dangerous operations (kill process, shutdown, delete) require confirmation

• Action history is logged for auditing

• Do not expose this server to the public internet without additional security measures

Tech Stack

• Backend: Python, Flask, Flask-RESTX, psutil

• Frontend: Vanilla JS, Chart.js, Lucide Icons, Inter Font

• AI Integration: MCP (Model Context Protocol), OpenAI Function Calling

• System: PowerShell, Windows API, pyautogui

License

MIT License - see LICENSE for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository

2. Create your feature branch (git checkout -b feature/amazing-feature)

3. Commit your changes (git commit -m 'Add amazing feature')

4. Push to the branch (git push origin feature/amazing-feature)

5. Open a Pull Request