This project provides an MCP (Model Context Protocol) server for interacting with Kaggle from Claude Desktop (or any MCP-compatible client) pip install -r requirements.txt Go to your Kaggle account -> Settings Scroll to API section → click Create New API Token. This will download a file called kaggle.json. Place kaggle.json in the following location on your system: ~/.kaggle/kaggle.json C:\Users\
Add this skill
npx mdskills install KrishnaPramodParupudi/kaggle-mcp-serverProvides Kaggle API integration but lacks detail on tools and over-declares permissions
1# kaggle-mcp-server2This project provides an MCP (Model Context Protocol) server for interacting with Kaggle from Claude Desktop (or any MCP-compatible client)34# 🚀 Setup5# 1. Clone the repository and install dependencies6 pip install -r requirements.txt78# 2. Configure Kaggle API credentials910 Go to your Kaggle account -> Settings1112 Scroll to API section → click Create New API Token.13 This will download a file called kaggle.json.1415 Place kaggle.json in the following location on your system:1617 Linux/Mac:1819 ~/.kaggle/kaggle.json2021 Windows:2223 C:\Users\<your-username>\.kaggle\kaggle.json2425# 3. Configure Claude Desktop2627 To connect Claude with this MCP server, create or update the file:2829 Claude/claude_desktop_config.json3031 with the following contents:3233 {34 "mcpServers": {35 "Kaggle": {36 "command": "<path-to-your-python-executable>",37 "args": ["<path-to-your-kaggle-server.py>"]38 }39 }40 }4142 Replace <path-to-your-python-executable> with the path to your Python interpreter (e.g. python or C:/Python/python.exe).4344 Replace <path-to-your-kaggle-server.py> with the path to this repo’s kaggle-server.py file.4546# 📜 Usage4748 Once configured:4950 Start Claude Desktop.5152 Start the Kaggle MCP server (Run kaggle-server.py as a Python File).5354 You can now use MCP tools defined in kaggle-server.py, for example:5556 get_competitions_list: Fetches the list of available Kaggle competitions and returns them in a JSON-friendly format.5758# 🛠️ About the Code5960 kaggle-server.py registers MCP tools using the @mcp.tool() decorator.6162 The server communicates with Claude via MCP and exposes Kaggle API methods.6364 Example tool:6566 @mcp.tool()67 def get_competitions_list() -> list[dict]:68 """69 Fetch the list of available Kaggle competitions.70 """7172# ✅ Requirements7374 Python 3.10+7576 Kaggle API credentials (kaggle.json)7778 Dependencies listed in requirements.txt7980# 🔒 Security Notes8182 Never commit your kaggle.json file.8384 Ensure the .kaggle folder has restricted permissions.85
Full transparency — inspect the skill content before installing.