{
  "markdown": "# Kaggle MCP Server\n\n<!-- mcp-name: io.github.Seif-Sameh/Kaggle-mcp -->\n\n[![PyPI](https://img.shields.io/pypi/v/mcp-server-kaggle.svg)](https://pypi.org/project/mcp-server-kaggle/)\n[![MCP Registry](https://img.shields.io/badge/MCP-registry-blue)](https://registry.modelcontextprotocol.io/v0.1/servers?search=io.github.Seif-Sameh/Kaggle-mcp)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)\n\nA Model Context Protocol (MCP) server that provides seamless integration with the Kaggle API. Interact with Kaggle competitions, datasets, kernels, and models through MCP-compatible clients like Claude Desktop.\n\n## Features\n\n- **Competitions**: List, download files, submit, view leaderboards and submissions\n- **Datasets**: Search, download, create, and manage datasets with version control\n- **Kernels**: List, push, pull, and manage Kaggle notebooks and scripts\n- **Models**: Create, update, and manage ML models and instances with full version control\n\n## Installation\n\n### Prerequisites\n\n- Python 3.10 or higher\n- A Kaggle account with API credentials\n\n### Install from PyPI\n\nThe recommended way is to run the server with [`uvx`](https://docs.astral.sh/uv/), which handles the install for you:\n\n```bash\nuvx mcp-server-kaggle\n```\n\nOr install it explicitly:\n\n```bash\npip install mcp-server-kaggle\n# or\nuv tool install mcp-server-kaggle\n```\n\n### Install from Source\n\nFor development or local modifications:\n\n```bash\ngit clone https://github.com/Seif-Sameh/Kaggle-mcp.git\ncd Kaggle-mcp\nuv sync\n```\n\n## Setup\n\n### 1. Get Your Kaggle API Credentials\n\n1. Go to [https://www.kaggle.com/account](https://www.kaggle.com/account)\n2. Scroll to the \"API\" section\n3. Click \"Create New Token\"\n4. This downloads `kaggle.json` with your credentials\n\n### 2. Configure Credentials\n\n**Option A: Environment Variables (Recommended)**\n\n```bash\nexport KAGGLE_USERNAME=your_username\nexport KAGGLE_API_KEY=your_api_key\n```\n\nOr add to your `~/.zshrc` or `~/.bashrc`:\n\n```bash\necho 'export KAGGLE_USERNAME=your_username' >> ~/.zshrc\necho 'export KAGGLE_API_KEY=your_api_key' >> ~/.zshrc\nsource ~/.zshrc\n```\n\n**Option B: Using .env File**\n\nCreate a `.env` file in your project directory:\n\n```env\nKAGGLE_USERNAME=your_username\nKAGGLE_API_KEY=your_api_key\n```\n\n## Usage\n\n### With Claude Desktop\n\nThe recommended way to use Kaggle MCP is with Claude Desktop.\n\n1. **Locate your Claude Desktop config file:**\n   - macOS: `~/Library/Application Support/Claude/claude_desktop_config.json`\n   - Windows: `%APPDATA%\\Claude\\claude_desktop_config.json`\n   - Linux: `~/.config/Claude/claude_desktop_config.json`\n\n2. **Add the Kaggle MCP server configuration:**\n\n```json\n{\n  \"mcpServers\": {\n    \"kaggle\": {\n      \"command\": \"uvx\",\n      \"args\": [\"mcp-server-kaggle\"],\n      \"env\": {\n        \"KAGGLE_USERNAME\": \"YOUR_KAGGLE_USERNAME\",\n        \"KAGGLE_API_KEY\": \"YOUR_KAGGLE_API_KEY\"\n      }\n    }\n  }\n}\n```\n\n<details>\n<summary>Running from a local source clone (alternative)</summary>\n\n```json\n{\n  \"mcpServers\": {\n    \"kaggle\": {\n      \"command\": \"uv\",\n      \"args\": [\n        \"--directory\",\n        \"/ABSOLUTE/PATH/TO/Kaggle-mcp\",\n        \"run\",\n        \"mcp-server-kaggle\"\n      ],\n      \"env\": {\n        \"KAGGLE_USERNAME\": \"YOUR_KAGGLE_USERNAME\",\n        \"KAGGLE_API_KEY\": \"YOUR_KAGGLE_API_KEY\"\n      }\n    }\n  }\n}\n```\n\n</details>\n\n3. **Restart Claude Desktop**\n\n4. **Start using Kaggle through Claude!**\n\nTry asking Claude:\n- \"List the latest Kaggle competitions\"\n- \"Download the Titanic dataset\"\n- \"Show me my recent competition submissions\"\n- \"Search for NLP datasets\"\n\n### Standalone Usage\n\nRun the MCP server directly:\n\n```bash\nmcp-server-kaggle\n```\n\nOr as a Python module:\n\n```bash\npython -m kaggle_mcp\n```\n\n\n## Available Tools\n\n### Competitions (8 tools)\n\n| Tool | Description |\n|------|-------------|\n| `competitions_list` | List and search available competitions |\n| `competition_list_files` | List all files in a competition |\n| `competition_download_file` | Download a specific competition file |\n| `competition_download_files` | Download all competition files |\n| `competition_submit` | Submit predictions to a competition |\n| `competition_submissions` | View your submission history |\n| `competition_leaderboard_view` | View the competition leaderboard |\n| `competition_leaderboard_download` | Download leaderboard data |\n\n### Datasets (10 tools)\n\n| Tool | Description |\n|------|-------------|\n| `datasets_list` | Search and filter datasets |\n| `dataset_metadata` | Get dataset metadata |\n| `dataset_list_files` | List files in a dataset |\n| `dataset_status` | Check dataset processing status |\n| `dataset_download_file` | Download a specific dataset file |\n| `dataset_download_files` | Download all dataset files |\n| `dataset_create` | Create a new dataset |\n| `dataset_initialize` | Initialize dataset metadata |\n| `dataset_create_version` | Create a new dataset version |\n\n### Kernels (7 tools)\n\n| Tool | Description |\n|------|-------------|\n| `kernels_list` | Search and filter kernels |\n| `kernel_list_files` | List files in a kernel |\n| `kernel_initialize` | Initialize kernel metadata |\n| `kernel_push` | Push a kernel to Kaggle |\n| `kernel_pull` | Download a kernel |\n| `kernel_output` | Download kernel output files |\n| `kernel_status` | Check kernel execution status |\n\n### Models (14 tools)\n\n| Tool | Description |\n|------|-------------|\n| `models_list` | Search and filter models |\n| `model_get` | Get model details and metadata |\n| `model_initialize` | Initialize model metadata |\n| `model_create` | Create a new model |\n| `model_update` | Update model information |\n| `model_delete` | Delete a model |\n| `model_instance_get` | Get model instance details |\n| `model_instance_initialize` | Initialize model instance metadata |\n| `model_instance_create` | Create a new model instance |\n| `model_instance_update` | Update a model instance |\n| `model_instance_delete` | Delete a model instance |\n| `model_instance_version_create` | Create a new model version |\n| `model_instance_version_download` | Download a model version |\n| `model_instance_version_delete` | Delete a model version |\n\n## Examples\n\n### Example 1: Working with Competitions\n\nAsk Claude:\n```\n\"List active Kaggle competitions about computer vision\"\n```\n\nClaude will use the `competitions_list` tool to search and display relevant competitions.\n\n### Example 2: Downloading Datasets\n\nAsk Claude:\n```\n\"Download the Titanic dataset to my Downloads folder\"\n```\n\nClaude will use `dataset_download_files` to fetch all dataset files.\n\n### Example 3: Submitting to Competitions\n\nAsk Claude:\n```\n\"Submit my predictions.csv to the Titanic competition with the message 'Initial baseline model'\"\n```\n\nClaude will use `competition_submit` to upload your submission.\n\n\n## License\n\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.\n",
  "bytes": 6836,
  "sha": "b1b64546db01df990da7b41a32ac4ed9e1af7577b2d6e9950be9a96769718322",
  "repo_slug": "seif-sameh/kaggle-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_seif_sameh_kaggle_mcp_0f19007f/readme"
}