Kaggle MCP Server
by Dishant27
The Kaggle MCP Server is a Model Context Protocol server designed to interact with Kaggle competitions through AI assistants like Claude. It allows users to perform various Kaggle-related tasks, such as listing competitions, downloading files, and submitting predictions, all through a conversational interface.
Last updated: N/A
What is Kaggle MCP Server?
The Kaggle MCP Server is a tool that enables interaction with Kaggle competitions using AI assistants. It acts as a bridge, allowing you to perform tasks like listing competitions, downloading data, and submitting predictions through a conversational interface.
How to use Kaggle MCP Server?
To use the server, you need to install Node.js, TypeScript, and the Kaggle CLI. After cloning the repository and installing dependencies, you can run the server and configure your AI assistant (e.g., Claude) to connect to it. Authentication with Kaggle API is required, either through a kaggle.json file or by setting environment variables in your AI assistant's configuration.
Key features of Kaggle MCP Server
List active Kaggle competitions
Search competitions by keyword
Download competition files
Submit prediction files to competitions
View submission history
Dataset Operations (Browse, search, download, and analyze datasets)
Notebook Integration (Work with Kaggle notebooks directly)
User Management (Check profile information and competition standings)
Competition Analysis (Get detailed competition metrics and leaderboard insights)
Use cases of Kaggle MCP Server
Quickly access competition details using natural language
Automate the process of downloading competition data
Submit predictions to competitions directly from an AI assistant
Monitor submission status and scores in real-time
Analyze datasets and notebooks from Kaggle using AI-powered tools
FAQ from Kaggle MCP Server
How do I authenticate with the Kaggle API?
How do I authenticate with the Kaggle API?
You can authenticate either by using a kaggle.json file in the ~/.kaggle/ directory or by setting the KAGGLE_USERNAME and KAGGLE_KEY environment variables in your AI assistant's configuration.
What if I encounter authentication errors?
What if I encounter authentication errors?
Verify that your credentials are correctly set and that the kaggle.json file has the correct permissions (600 on Linux/Mac). Also, ensure that your API token is still valid.
How do I install the Kaggle CLI?
How do I install the Kaggle CLI?
You can install the Kaggle CLI using pip: pip install kaggle. Make sure the Kaggle command is in your PATH.
Are all features available?
Are all features available?
Yes, all features are now available in the feature-complete branch of the repository. This includes dataset operations, notebook integration, user management, and competition analysis.
How do I run the server?
How do I run the server?
After installing dependencies and building the project, you can run the server using npm start or directly with node build/index.js.