kaggle-mcp

kaggle-mcp

6
research_and_dataai_chatbot

This project implements a Model Context Protocol server for interacting with the Kaggle API, allowing searches and downloads of Kaggle datasets and prompts for creating EDA notebooks. It simplifies data handling processes for developers by automating these tasks through a server application.

Kaggle MCP (Model Context Protocol) Server

This project provides a Model Context Protocol server built using the FastMCP library to interface with the Kaggle API. It allows users to search for and download datasets and generate prompts for Exploratory Data Analysis (EDA) notebooks.

Project Structure

  • Defines practical resources, tools, and prompts for Kaggle interaction.
  • Uses Python for server deployment and environment setups.

Server Features

  • search_kaggle_datasets: Look for datasets based on a query.
  • download_kaggle_dataset: Download datasets by reference.
  • generate_eda_notebook: Produce EDA notebook prompts for datasets.

Running the Server

  • Activate a virtual environment and execute the server using uv commands. Ensure Kaggle API credentials are set in environment variables or kaggle.json file.

Docker Utilization

  • Build Docker images and run containers using environment variables for seamless deployment.