kaggle-mcp
7
Kaggle-MCP is a tool that integrates Claude AI with the Kaggle API, enabling AI-powered access to Kaggle competitions, datasets, kernels, and models through the Model Context Protocol. It facilitates secure authentication and data access for research and analysis tasks.
Kaggle-MCP: Kaggle API Integration for Claude AI
Kaggle-MCP connects Claude AI with the Kaggle API via the Model Context Protocol (MCP), allowing AI-driven operations on competitions, datasets, kernels, and models.
Features
- Authentication: Securely authenticate with Kaggle credentials
- Competitions: Browse and download competition data
- Datasets: Discover and download Kaggle datasets
- Kernels: Explore and analyze Kaggle notebooks
- Models: Access pre-trained models from Kaggle
Quick Installation
- macOS/Linux: Use the provided shell command for installation
- Windows: Follow the PowerShell command instructions for setup
- Manual: Use pip or uv pip for installation
Configuration
Run kaggle-mcp-setup
to configure Claude Desktop or manually add configuration to your desktop configuration file.
Kaggle API Credentials
Set up your Kaggle API credentials by downloading the kaggle.json
file from your account settings and placing it in the correct directory.
Use Cases
- Competition Research: Access details and leaderboards
- Dataset Discovery: Find datasets for projects
- Learning Resources: Locate relevant kernels
- Model Discovery: Explore pre-trained models
Requirements
- Python 3.8+
- Claude Desktop or API access
- Kaggle account with API credentials
- MCP Python SDK 1.6.0+