Mcpcontrol

MCPControl is a Windows system control server that provides programming control for system operations such as mouse, keyboard, window management, and screen capture through the Model Context Protocol.

Operating system automation Developer tools #System Control #Windows Tools #Automated Operations #Screen Management .TypeScript

rating : 2.5 points

downloads : 6.6K

update time : 2025-04-29

Open Site

What is MCPControl?

MCPControl is a Windows-based control server that allows you to operate your computer programmatically, such as moving the mouse, entering text, and adjusting window positions.

How to use MCPControl?

You can easily call various functions of MCPControl through simple JSON-RPC requests. Just start the server and send commands to begin using it.

Applicable scenarios

MCPControl is very suitable for developers who need to automate tasks or users who want to control their computers programmatically.

Main Features

Window Management

List all windows, get information about the active window, set the size and position of windows, etc.

Mouse Control

Move the mouse, click, drag, and track the mouse position.

Keyboard Control

Enter text, perform key combination operations, press or release individual keys.

Screen Capture

Capture screenshots of the entire screen or specific windows.

Clipboard Integration

Read, write, and clear the clipboard content.

Advantages

Supports various system-level operations, such as window management and clipboard control.

Cross-platform compatibility (currently only for Windows).

Easy to integrate into existing projects.

Limitations

Some functions may require administrator privileges.

Does not support complex operations on multi-screen devices.

Some key combinations may cause crashes (to be fixed soon).

How to Use

Install Dependencies

Make sure Node.js and necessary build tools are installed.

Start the Server

Run npx mcp-control to start the MCPControl service.

Send Requests

Use JSON-RPC to send commands to perform the desired operations.

Usage Examples

Get a Screen Screenshot

Get the current screen screenshot by calling get_screenshot.

Simulate a Mouse Click

Use click_mouse to click at a certain position on the screen.

Frequently Asked Questions

Does MCPControl support Linux or macOS?

Why do some key combinations cause crashes?

How to stop the MCPControl service?

Related Resources

Official Documentation

Complete documentation for the MCP specification.

GitHub Repository

Source code and issue tracking for the MCPControl project.

Example Scripts

Some useful MCPControl example scripts.

🚀 MCPControl Server

MCPControl is a control server based on the Windows model context protocol, offering programmatic control over system operations, including mouse, keyboard, window management, and screen capture capabilities.

🚀 Quick Start

🎥 Quick Demo (30-Second Stunning Demo)

Want to see what MCPControl can do? Try our interactive demo:

# Run the demo with a single command
node demo.cjs

The demo will:

Showcase available MCPControl tools
Display your screen size
Track the mouse position
Retrieve active window information
Optionally capture screenshots and save them to your desktop

No need to run additional commands; the demo manages the MCPControl server automatically.

⚠️ Important Note

This project currently only supports the Windows operating system.

⚠️ Important Disclaimer

This software is experimental and potentially dangerous.

By using this software, you acknowledge and accept the following:

There are inherent risks in allowing an AI model to directly control your computer through this tool.
This software can control the mouse, keyboard, and other system functions, which may lead to unexpected consequences.
You use this software entirely at your own risk.
The creators and contributors of this project are not responsible for any damage, data loss, or other consequences resulting from the use of this software.
This tool should be used in a controlled environment with appropriate security measures.

Use at your own risk.

✨ Features

Window Management
- List all windows
- Get active window information
- Get window titles
- Get window size and position
- Focus on windows
- Resize windows
- Move windows
Mouse Control
- Move the mouse
- Perform click operations
- Scroll functionality
- Drag and drop operations
- Track mouse position
Keyboard Control
- Text input
- Key combinations
- Key press/release operations
- Lock key functionality
Screen Operations
- Screen capture
- Get screen size
- Detect active windows
Clipboard Integration
- Get clipboard content
- Set clipboard content
- Clear the clipboard
- Get clipboard status

And many more!

🔧 Known Limitations

Window minimize/restore operations are currently not supported.
Multi-screen functionality may not work correctly depending on settings.
The get_screenshot utility is incompatible with the VS Code extension Cline. See GitHub Issue #1865.
Some operations may require elevated privileges, depending on the target application.
Only the Windows operating system is supported.
Ctrl key combinations (e.g., Ctrl+C, Ctrl+V) may cause the server to crash due to stdio handling issues. This will be addressed in an upcoming release with a new streaming HTTP transport protocol in the MCP specification. See GitHub Issue #120.