Transform your Windows PC into an intelligent automation hub controlled by AI assistants like Claude, ChatGPT, and more.
MCP Windows Server is an AI-native automation framework that enables AI assistants to control Windows systems through natural language commands. Built on the Model Context Protocol (MCP), it provides secure, comprehensive system-level automation capabilities.
MCP is an open protocol by Anthropic that allows AI models to interact safely with local tools, APIs, and system services. Our server implements this protocol for Windows, making AI assistants powerful desktop automation agents.
- π Secure System Access - Controlled command execution with safety filters
- π§ AI Agent Compatible - Works with Claude, ChatGPT, and other AI assistants
- π Real-Time Communication - Instant bidirectional AI
βοΈ System interaction - π§ Plugin Architecture - Extensible framework for custom automation
# Install from PyPI
pip install mcp-windows-server
# Or install from source
git clone https://github.com/Mahipal/mcp-windows-server.git
cd mcp-windows-server
pip install -r requirements.txt
# Start the MCP server
mcp-windows-server
# Or use the unified server directly
unified-server
- Install the package:
pip install mcp-windows-server
- Configure Claude Desktop:
{
"mcpServers": {
"mcp-windows-server": {
"command": "mcp-windows-server",
"env": {
"PYTHONUNBUFFERED": "1"
}
}
}
}
- Restart Claude Desktop and start automating!
- β 200+ Automation Tools - Comprehensive Windows control
- π§ AI-Context Aware - Understands natural language commands
- π Bidirectional Communication - Real-time AI
βοΈ System interaction - βοΈ Safe Execution - Built-in command filtering and validation
- π§± Modular Design - Plugin-based architecture
- π§ͺ ML Integration - Machine learning for predictive automation
Category | Description | Examples |
---|---|---|
π₯οΈ System Control | Process, registry, services management | Kill processes, manage services, registry edits |
π File Operations | File system automation | Copy, move, search, backup files |
π Web Automation | Browser control and web scraping | Form filling, data extraction, navigation |
πΌοΈ Image Processing | Screenshot and image manipulation | OCR, image editing, screen capture |
π Data Analysis | ML training and data processing | Model training, data visualization |
π’ Office Integration | Microsoft Office automation | Excel reports, Word documents, PowerPoint |
π Security | System security and monitoring | Firewall rules, security scans, monitoring |
π Network | Network configuration and monitoring | WiFi management, network diagnostics |
Once integrated with Claude Desktop, you can use natural language:
"Take a screenshot and save it as desktop.png"
"Get my system information"
"List all running processes"
"Create a backup of my Documents folder"
"Check my network connection status"
"Open Calculator application"
mcp-windows-server/
βββ mcp_windows_automation/ # Main package
β βββ __init__.py
βββ unified_server.py # Core MCP server
βββ office_mcp_server.py # Office integration
βββ config/ # Configuration templates
βββ docs/ # Documentation
βββ examples/ # Usage examples
βββ requirements.txt # Dependencies
βββ README.md # This file
# Optional MySQL database integration
MYSQL_USER=your_username
MYSQL_PASSWORD=your_password
MYSQL_DATABASE=your_database
MYSQL_HOST=localhost
MYSQL_PORT=3306
# Python configuration
PYTHONPATH=/path/to/mcp-windows-server
PYTHONUNBUFFERED=1
For advanced setups, copy the configuration template:
cp config/claude_desktop_config.template.json config/claude_desktop_config.json
# Edit with your specific paths and credentials
git clone https://github.com/Mahipal/mcp-windows-server.git
cd mcp-windows-server
pip install -r requirements.txt
python unified_server.py
python -m build
pip install dist/mcp_windows_server-*.whl
- Command Filtering: Dangerous commands are blocked by default
- Safe Execution: All operations run in controlled environment
- No Credential Storage: Sensitive data excluded from package
- Template-Based Config: Only safe configuration templates included
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature
) - Commit your changes (
git commit -m 'Add amazing feature'
) - Push to the branch (
git push origin feature/amazing-feature
) - Open a Pull Request
This project is licensed under the MIT License - see the LICENSE file for details.
- Author: Mahipal
- Email: mukuljangra5@gmail.com
- GitHub: Mahipal
- PyPI: mcp-windows-server
- Anthropic for the Model Context Protocol
- Claude for AI assistant integration
- The open-source community for inspiration and contributions
"Automate Everything. With AI." π§ π»
Made with β€οΈ for the AI automation community