What is Kaggle-MCP?
Kaggle-MCP is a tool that connects Claude AI to the Kaggle API via the Model Context Protocol (MCP), enabling seamless operations with competitions, datasets, and kernels.
How do I install Kaggle-MCP on macOS or Linux?
You can install Kaggle-MCP on macOS or Linux by running the command: curl -LsSf https://raw.githubusercontent.com/54yyyu/kaggle-mcp/main/install.sh | sh
What do I need to use Kaggle-MCP?
To use Kaggle-MCP, you need Python 3.8 or newer, Claude Desktop or API access, a Kaggle account with API credentials, and the MCP Python SDK 1.6.0+.
How do I authenticate Kaggle-MCP?
Authenticate Kaggle-MCP by setting up your Kaggle API credentials. Download the kaggle.json
file from your Kaggle account settings and move it to ~/.kaggle/kaggle.json
.
Can I manually configure Kaggle-MCP?
Yes, you can manually add the necessary configuration to your Claude Desktop configuration file to set up Kaggle-MCP.
Kaggle-MCP
Project Details
- 54yyyu/kaggle-mcp
- MIT License
- Last Updated: 4/13/2025
Recomended MCP Servers
Chronos – A Stellar MCP server
MCP tool that lets Cline inquire about a code base
MCP server for Apache Gravitino(incubating)
Git stuff MCP server
StarRocks MCP (Model Context Protocol) Server
MCP Server example with TypeScript