Computer Control MCP

🚀 Computer Control MCP Server

A MCP server that provides computer control functions such as mouse, keyboard, OCR, etc., using PyAutoGUI, RapidOCR, and ONNXRuntime. Similar to Anthropic's 'computer-use'. No external dependencies.

⚠️ Important Note

This project has only been tested on Windows. It may work on other platforms.

🚀 Quick Start

Install MCP via uvx

{
  "mcpServers": {
    "computer-control-mcp": {
      "command": "uvx",
      "args": ["computer-control-mcp@latest"]
    }
  }
}

Install Globally via pip

pip install computer-control-mcp

Run the Server

computer-control-mcp # Instead of using uvx computer-control-mcp, this way you can use the latest version. You can also use `uv cache clean` to clear the cache and then use `uvx` again to get the latest version.

✨ Features

• Control mouse movement and clicks.
• Input text at the current cursor position.
• Take screenshots of the entire screen or specific windows, optionally save them to the download directory.
• Extract text from screenshots using OCR (Optical Character Recognition).
• List and activate windows.
• Press keyboard keys.
• Perform drag-and-drop operations.

📦 Available Tools

Mouse Control

• click_screen(x: int, y: int): Click at the specified screen coordinates.
• move_mouse(x: int, y: int): Move the mouse to the specified coordinates.
• drag_mouse(from_x: int, from_y: int, to_x: int, to_y: int, duration: float = 0.5): Drag the mouse from one position to another.

Keyboard Control

• type_text(text: str): Input the specified text at the current cursor position.
• press_key(key: str): Press the specified keyboard key.

Screen and Window Management

• take_screenshot(title_pattern: str = None, use_regex: bool = False, threshold: int = 60, with_ocr_text_and_coords: bool = False, scale_percent_for_ocr: int = 100, save_to_downloads: bool = False): Take a screenshot of the screen or window, optionally with OCR.
• get_screen_size(): Get the current screen resolution.
• list_windows(): List all open windows.
• activate_window(title_pattern: str, use_regex: bool = False, threshold: int = 60): Activate the specified window.

🛠️ Development Environment Setup

1. Clone the repository:

git clone https://github.com/AB498/computer-control-mcp.git
cd computer-control-mcp

2. Install dependencies:

pip install -r requirements.txt

3. Start the server:

python server.py

🧪 Run Tests

pytest tests/

📚 API Reference

For the full API documentation, please visit: https://github.com/AB498/computer-control-mcp/blob/main/docs/api.md

📄 License

MIT License

📞 Contact

If you need help or feedback, please contact:

• Email: [Insert email address here]
• Discord: [Insert Discord invite link here]