MCP Catalogs
← Home

kaggle-MCP

by Dishant27·★ 9·Score 34

MCP server enabling AI assistants to interact with Kaggle competitions through commands and API access.

ai-llmdeveloper-toolsdata-science
2
Forks
2
Open issues
14 mo ago
Last commit
2d ago
Indexed

Overview

The Kaggle MCP server provides a comprehensive interface between AI assistants and the Kaggle platform. It allows users to browse competitions, download datasets, submit predictions, and track submissions all through natural language commands to compatible AI systems. The server supports both kaggle.json file authentication and environment variable configuration, making it flexible for different setup preferences.

Try asking AI

After installing, here are 6 things you can ask your AI assistant:

you:Data scientists can quickly download competition datasets without leaving their AI coding environment
you:AI assistants can help prepare and submit competition entries through natural language commands
you:Researchers can stay updated on Kaggle competitions and track their submission history
you:How do I authenticate with Kaggle?
you:What competitions can I access with this server?
you:Can I submit predictions to competitions?

When to choose this

Choose this server if you're a data scientist or ML practitioner who wants to integrate Kaggle competitions directly into your AI workflow without switching contexts between the Kaggle website and your AI assistant.

When NOT to choose this

Don't choose this if you need access to private competitions (requires manual acceptance of terms on Kaggle) or if you're already committed to another data science platform with built-in AI integration.

Tools this server exposes

7 tools extracted from the README
  • list_competitions

    List all active Kaggle competitions or search competitions by keyword

  • download_competition_files

    Download files for any Kaggle competition

  • submit_to_competition

    Submit prediction files to competitions with custom messages

  • list_submissions

    View submission history for a specific competition

  • browse_datasets

    Browse and search datasets from Kaggle

  • analyze_competition

    Get detailed competition metrics and leaderboard insights

  • manage_user_profile

    Check profile information and competition standings

Comparable tools

kaggle-apimcp-server-kaggledata-science-workflow-mcp

Installation

Installation

  1. Install prerequisites:

```bash # Install Kaggle CLI pip install kaggle

# Install Node.js 16+ and TypeScript # (follow instructions for your OS) ```

  1. Get Kaggle API credentials:

- Go to https://www.kaggle.com/account and click "Create New API Token" - This will download a kaggle.json file - Place it in ~/.kaggle/ directory (Linux/Mac) or %USERPROFILE%\.kaggle\ (Windows)

  1. Install the MCP server:

``bash git clone https://github.com/Dishant27/kaggle-MCP.git cd kaggle-MCP npm install npm run build ``

  1. Configure Claude Desktop:

``json { "mcpServers": { "kaggle": { "command": "node", "args": ["/path/to/kaggle-MCP/build/index.js"], "env": { "KAGGLE_USERNAME": "your-kaggle-username", "KAGGLE_KEY": "your-kaggle-api-key" } } } } ``

FAQ

How do I authenticate with Kaggle?
You can use either a kaggle.json file in your ~/.kaggle directory or configure your credentials in the env section of your claude_desktop_config.json file.
What competitions can I access with this server?
The server provides access to all public Kaggle competitions. You can list competitions, search by keyword, and download files for any public competition.
Can I submit predictions to competitions?
Yes, the server allows you to submit prediction files to competitions with custom messages and get immediate feedback on submission status.

Compare kaggle-MCP with

kaggle-MCP vs ultimate_mcp_serverkaggle-MCP vs mcp-server-chartkaggle-MCP vs everythingkaggle-MCP vs filesystemkaggle-MCP vs time
GitHub →

Last updated · Auto-generated from public README + GitHub signals.