Quick Data for Windows MCP

Windows-optimized fork of disler/quick-data-mcp for Claude Desktop

Universal data analytics capabilities for JSON/CSV files - now working seamlessly on Windows!

🚀 What This Does

This is a Windows-optimized fork of the excellent quick-data-mcp project by @disler.

The original project provides powerful MCP server capabilities for data analytics, and this fork specifically addresses Windows compatibility issues and Claude Desktop integration challenges.

**This is my first ever try at this. Please feel free to give suggestions and or criticisms. I loved the quick data mcp for claude code. There was nothing available like it for claude desktop so with the help of claude code we now have it.

Key Improvements Over Original:

• ✅ Windows Path Handling - Proper Windows file path support
• ✅ Claude Desktop Ready - Pre-configured batch launchers and setup
• ✅ Dependency Management - Automated installation scripts
• ✅ Troubleshooting - Complete guides for common Windows issues

✨ Key Features

• Universal Data Support - Works with any CSV/JSON file structure
• Windows Path Optimization - Handles Windows file paths correctly
• Claude Desktop Integration - Pre-configured for seamless setup
• Automatic Schema Discovery - Analyzes your data and suggests analyses
• 32+ Analytics Tools - From basic stats to advanced ML features
• Interactive Visualizations - Create charts with Plotly
• Memory Management - Optimized for large datasets

🏁 Quick Start for Windows

Prerequisites

• Windows 10/11
• Python 3.9+ (Download here)
• Claude Desktop (Download here)

Installation

1. Download or clone this repository:

   git clone https://github.com/Beaulewis1977/quick-data-for-windows-mcp.git
   cd quick-data-for-windows-mcp
   

2. Install dependencies:

   install_dependencies.bat
   

3. Test the server:

   test_server.bat
   

4. Configure Claude Desktop:

   Copy the fixed configuration to Claude Desktop:

   copy claude_desktop_config_fixed.json "%APPDATA%\Claude\claude_desktop_config.json"
   

   IMPORTANT: Edit the config file and update the cwd path to your actual installation directory.

5. Restart Claude Desktop

🚨 Having Issues?

If you see ModuleNotFoundError: No module named 'mcp', check the TROUBLESHOOTING.md guide.

💻 Usage in Claude Desktop

Once configured, start with this slash command in Claude Desktop:

/quick-data-windows

Loading Your Data

Load my sales data: C:\Users\YourName\Documents\sales_data.csv as "sales"

Basic Analysis

Show me correlations in the sales dataset
Create a bar chart of sales by region
Analyze the distribution of revenue column

Advanced Analytics

Validate data quality for sales dataset
Compare sales dataset with marketing dataset
Generate dashboard with revenue trends and regional breakdown

🔧 Available Tools

Dataset Management

• load_dataset - Load CSV/JSON files with automatic schema discovery
• list_loaded_datasets - View all datasets in memory
• get_dataset_info - Get detailed dataset information
• clear_dataset / clear_all_datasets - Memory management

Core Analytics

• segment_by_column - Analyze categorical data segments
• find_correlations - Discover relationships between variables
• analyze_distributions - Statistical distribution analysis
• detect_outliers - Identify data anomalies
• suggest_analysis - AI-powered analysis recommendations

Visualization

• create_chart - Generate interactive charts (bar, scatter, line, histogram)
• generate_dashboard - Multi-chart dashboards

Advanced Analytics

• validate_data_quality - Comprehensive data quality scoring
• compare_datasets - Multi-dataset comparison analysis
• merge_datasets - Join datasets with flexible strategies
• calculate_feature_importance - ML feature importance analysis
• export_insights - Export results in multiple formats

📂 Supported File Formats

CSV Files

• Standard CSV with headers
• Custom delimiters automatically detected
• UTF-8 encoding support
• Large file handling with sampling options

JSON Files

• Flat JSON structures
• Nested JSON (automatically flattened)
• JSON Lines format
• Array of objects format

🛠️ Configuration

Manual Configuration

If the automatic setup doesn't work, manually edit your Claude Desktop config:

Location: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "quick-data-windows": {
      "command": "python",
      "args": [
        "C:\\path\\to\\your\\quick-data-for-windows-mcp\\main.py"
      ],
      "cwd": "C:\\path\\to\\your\\quick-data-for-windows-mcp",
      "env": {
        "LOG_LEVEL": "INFO",
        "PYTHONPATH": "C:\\path\\to\\your\\quick-data-for-windows-mcp\\src"
      }
    }
  }
}

Alternative: Using UV Package Manager

If you prefer UV (recommended for Python dependency management):

{
  "mcpServers": {
    "quick-data-windows": {
      "command": "uv",
      "args": [
        "--directory",
        "C:\\path\\to\\your\\quick-data-for-windows-mcp",
        "run",
        "python",
        "main.py"
      ]
    }
  }
}

🧪 Testing the Server

Test the server standalone (before Claude Desktop integration):

python main.py

Expected output:

Quick Data for Windows MCP v1.0.0
Server running on stdio...

📊 Example Workflows

Sales Data Analysis

1. Load sales_data.csv as "sales"
2. Show correlations in sales dataset
3. Create bar chart of sales by product_category 
4. Detect outliers in revenue column
5. Generate dashboard with top products and regional trends

Data Quality Assessment

1. Load customer_data.csv as "customers"
2. Validate data quality for customers dataset
3. Analyze distributions for age column
4. Segment by customer_type column

🔍 Troubleshooting

Common Issues

"Module not found" errors:

• Ensure Python is in your PATH
• Run pip install -r requirements.txt manually
• Check that PYTHONPATH is set correctly in config

"File not found" errors:

• Use full Windows paths: C:\Users\...
• Avoid relative paths like .\data\file.csv
• Check file permissions

Claude Desktop not finding server:

• Restart Claude Desktop after config changes
• Check config file syntax with JSON validator
• Verify file paths are correct (no typos)

Getting Help

1. Check that Python 3.9+ is installed: python --version
2. Verify dependencies: pip list | findstr pandas
3. Test server manually: python main.py
4. Check Claude Desktop logs for errors

🤝 Contributing

This is a community-driven Windows adaptation of the original quick-data-mcp project. Contributions welcome!

Development Setup

# Clone and setup
git clone https://github.com/Beaulewis1977/quick-data-for-windows-mcp.git
cd quick-data-for-windows-mcp

# Install development dependencies
pip install -r requirements.txt
pip install pytest black ruff

# Run tests (when implemented)
pytest tests/

📝 License

MIT License - see LICENSE file for details.

🙏 Acknowledgments

This project is a Windows-optimized fork of the original quick-data-mcp by @disler.

Original Project Credits

• Original Author: @disler
• Original Repository: disler/quick-data-mcp
• Original Purpose: MCP server for data analytics with Claude Code
• License: MIT (maintained in this fork)

Windows Fork Contributions

• Windows Compatibility: @Beaulewis1977
• Claude Desktop Integration: Community-driven improvements
• Troubleshooting & Documentation: Enhanced for Windows users

Technology Stack

• Model Context Protocol: Anthropic
• Data Processing: pandas, numpy, plotly, scikit-learn
• Platform: Optimized for Windows + Claude Desktop

⭐ Please star both repositories:

• Original Project - For the core innovation
• This Fork - For Windows support

Special thanks to @disler for creating the foundational work that made this Windows adaptation possible!

🔗 Links

• Claude Desktop
• Model Context Protocol
• Original Quick Data MCP
• Python for Windows

---

**This is my first ever try at this. Please feel free to give suggestions and or criticisms. I loved the quick data mcp for claude code. There was nothing available like it for claude desktop so with the help of claude code we now have it.

Ready to analyze your data with AI? Load a CSV and start exploring! 🚀