MCP Server for Data Exploration

MCP Server is a versatile tool designed for interactive data exploration.

Your personal Data Scientist assistant, turning complex datasets into clear, actionable insights.

🚀 Try it Out

1. Download Claude Desktop

   • Get it here

2. Install and Set Up

   • On macOS, run the following command in your terminal:

   python setup.py

3. Load Templates and Tools

   • Once the server is running, wait for the prompt template and tools to load in Claude Desktop.

4. Start Exploring

   • Select the explore-data prompt template from MCP
   • Begin your conversation by providing the required inputs:
     • csv_path: Local path to the CSV file
     • topic: The topic of exploration (e.g., "Weather patterns in New York" or "Housing prices in California")

Examples

These are examples of how you can use MCP Server to explore data without any human intervention.

Case 1: California Real Estate Listing Prices

• Kaggle Dataset: USA Real Estate Dataset
• Size: 2,226,382 entries (178.9 MB)
• Topic: Housing price trends in California

• Data Exploration Summary

Case 2: Weather in London

• Kaggle Dataset: 2M+ Daily Weather History UK
• Size: 2,836,186 entries (169.3 MB)
• Topic: Weather in London
• Report: View Report
• Graphs:
  • London Temperature Trends

• Temperature-Humidity Relationship by Season

• Wind Direction Pattern by Season

📦 Components

Prompts

• explore-data: Tailored for data exploration tasks

Tools

1. load-csv

   • Function: Loads a CSV file into a DataFrame
   • Arguments:
     • csv_path (string, required): Path to the CSV file
     • df_name (string, optional): Name for the DataFrame. Defaults to df_1, df_2, etc., if not provided

2. run-script

   • Function: Executes a Python script
   • Arguments:
     • script (string, required): The script to execute

⚙️ Modifying the Server

Claude Desktop Configurations

• macOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%/Claude/claude_desktop_config.json

Development (Unpublished Servers)

"mcpServers": {
  "mcp-server-ds": {
    "command": "uv",
    "args": [
      "--directory",
      "/Users/username/src/mcp-server-ds",
      "run",
      "mcp-server-ds"
    ]
  }
}

Published Servers

"mcpServers": {
  "mcp-server-ds": {
    "command": "uvx",
    "args": [
      "mcp-server-ds"
    ]
  }
}

🛠️ Development

Building and Publishing

1. Sync Dependencies

   uv sync

2. Build Distributions

   uv build

   Generates source and wheel distributions in the dist/ directory.

3. Publish to PyPI

   uv publish

🤝 Contributing

Contributions are welcome! Whether you're fixing bugs, adding features, or improving documentation, your help makes this project better.

Reporting Issues

If you encounter bugs or have suggestions, open an issue in the issues section. Include:

• Steps to reproduce (if applicable)
• Expected vs. actual behavior
• Screenshots or error logs (if relevant)

📜 License

This project is licensed under the MIT License. See the LICENSE file for details.

💬 Get in Touch

Questions? Feedback? Open an issue or reach out to the maintainers. Let's make this project awesome together!

About

This is an open source project run by ReadingPlus.AI LLC. and open to contributions from the entire community.