Content
# Claude MCP Server Collection
This repository contains a collection of Model Context Protocol (MCP) servers designed to enhance Claude's desktop application capabilities. Each server provides specific functionality that allows Claude to interact with your computer in different ways.
## Latest Addition: DuckDB Integration! 🚀
New DuckDB FastAPI server for efficient large-scale data analysis:
### Example Queries and Results
#### Query 1: Analysis
#### Query 2: Aggregated Results
### DuckDB Server Features
- Analyze multiple CSV files (>1GB) directly in Claude Desktop
- Lightning-fast SQL queries with DuckDB
- Memory-efficient data processing
- Connection pooling and caching
- Automatic cleanup of unused connections
## Overview
The project consists of several MCP servers:
1. Screen Capture Server - Captures and processes screenshots
2. Computer Control Server - Enables keyboard and mouse automation
3. FastAPI Integration Server - Handles data processing and API endpoints
4. Curl Server - Provides HTTP request capabilities
5. DuckDB Server - Enables large-scale data analysis
## Prerequisites
- Python 3.8 or higher
- Node.js and npm (for filesystem server)
- Claude Desktop Application
- Git
## Installation
1. Clone the repository:
```bash
git clone https://github.com/syedazharmbnr1/ClaudeMCPServer.git
cd ClaudeMCPServer
```
2. Create and activate a virtual environment:
```bash
python3 -m venv .env
source .env/bin/activate # On Windows: .env\Scripts\activate
```
3. Install dependencies:
```bash
pip install -r requirements.txt
```
## Server Components
### 1. DuckDB Server
Enables Claude to analyze large CSV files efficiently.
#### Setup and Usage:
```bash
cd fastapi/duckdb
python main.py
```
#### Features:
- Process multiple CSV files simultaneously
- Handle files larger than 1GB
- Fast SQL querying with DuckDB
- Memory-efficient processing
- Connection pooling
### 2. Screen Capture Server
Enables Claude to capture and process screenshots of your screen.
#### Setup and Usage:
```bash
python mcp_screen_server.py
```
#### Features:
- Real-time screen capture
- Dynamic image compression
- WebP format support for optimal file size
- Customizable save locations
### 3. Computer Control Server
Allows Claude to control mouse and keyboard actions.
#### Setup and Usage:
```bash
python ComputerUse/mcp_computer_server.py
```
#### Features:
- Mouse movement and clicks
- Keyboard shortcuts and text input
- Screen position tracking
- Clipboard operations
### 4. FastAPI Integration Server
The FastAPI server provides a robust API interface for data processing and integration.
#### Setup and Configuration:
1. Navigate to the FastAPI directory:
```bash
cd fastapi
```
2. Configure environment variables:
```bash
export PYTHONPATH=/path/to/mcp-server-py
export PORT=8000
```
3. Start the server:
```bash
python main.py
```
### 5. Curl Server
Provides HTTP request capabilities to Claude.
#### Setup:
```bash
cd Curl_Server
./start_server.sh # For the basic server
./start_mcp_server.sh # For the MCP integration
```
## Claude Desktop Integration
### Configuration
1. Copy the `claude_desktop_config.json` to your Claude Desktop app configuration directory
2. Update the paths in the configuration to match your system:
```json
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-filesystem",
"<your-paths-here>"
]
},
"duckdb": {
"command": "/path/to/python",
"args": [
"/path/to/fastapi/duckdb/main.py"
],
"cwd": "/path/to/fastapi/duckdb",
"env": {
"PYTHONPATH": "/path/to/mcp-server-py",
"PORT": "8010"
}
}
}
}
```
## Testing
Each component has its own test suite:
```bash
# Test screen capture
python test_screen_server.py
# Test computer control
python ComputerUse/test_computer_control.py
# Test integration
python ComputerUse/test_client.py
```
## Troubleshooting
### Common Issues
1. Python Path Issues:
- Ensure PYTHONPATH is set correctly
- Verify virtual environment is activated
2. Permission Errors:
- Make sure script files are executable:
```bash
chmod +x *.py
chmod +x Curl_Server/*.sh
```
3. Port Conflicts:
- Screen Server: Default port 8767
- FastAPI Server: Default port 8000
- DuckDB Server: Default port 8010
- Integration Server: Default port 8768
### Logging
- Check `debug.log` for detailed error messages
- Each server component writes to its own log file
## Security Notes
1. GitHub Integration:
- Store your GitHub token securely
- Never commit tokens to the repository
2. File System Access:
- Configure filesystem paths carefully
- Limit access to necessary directories only
## Contributing
1. Fork the repository
2. Create a feature branch
3. Submit a pull request
## License
This project is licensed under the MIT License - see the LICENSE file for details.
MCP Store