kaggle-mcp-server
Provides comprehensive tools for interacting with Kaggle's API, enabling AI agents to manage competitions, datasets, kernels/models, discussions, and benchmarks programmatically.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Kaggle MCP Server
A full-featured Model Context Protocol (MCP) server for the Kaggle API — 96 tools across competitions, datasets, kernels, models, benchmarks, discussions, and workflow utilities.
Installation
pip install kaggle-mcp-serverPrerequisites
Kaggle API credentials — place your
kaggle.jsonat~/.kaggle/kaggle.json:
{"username":"YOUR_USERNAME","key":"YOUR_API_KEY"}Get your API key from kaggle.com/settings → "Create New Token".
Python 3.12+
Related MCP server: kaggle-mcp
Usage
With Cursor IDE
Add to .cursor/mcp.json:
{
"mcpServers": {
"kaggle": {
"command": "kaggle-mcp-server"
}
}
}With Claude Desktop
Add to your Claude Desktop config:
{
"mcpServers": {
"kaggle": {
"command": "kaggle-mcp-server"
}
}
}Standalone
kaggle-mcp-serverTools (96 total)
Competitions (16 tools)
Tool | Description |
| Search and list competitions |
| Get detailed competition info |
| List competition data files |
| Hierarchical tree view of competition data |
| Download competition data (returns URL) |
| Download a single competition file locally |
| Submit predictions via blob token |
| Submit a local prediction file |
| Submit to code competitions |
| View submission history |
| Get single submission details |
| Get/poll submission score |
| View top 20 leaderboard |
| Download full leaderboard |
| Find a team/user's rank |
| Get data files summary |
Competition Workflow (5 tools)
Tool | Description |
| Download and extract competition data locally |
| One-shot setup: info + download + preview |
| Show competitions with closest deadlines |
| List competitions you've entered |
| Top public notebooks for a competition |
Datasets (20 tools)
Tool | Description |
| Search and list datasets |
| Get full dataset info |
| List files in a dataset |
| Hierarchical tree view of dataset files |
| Get file count and total size |
| Download dataset (returns URL) |
| Download a single dataset file |
| Download and extract dataset locally |
| Get dataset metadata |
| Update title/description/license |
| Create dataset via blob tokens |
| Create dataset from local directory |
| Create new version via tokens |
| Push new version from local directory |
| Delete a dataset |
| Check dataset processing status |
| Upload file and get blob token |
| List your datasets |
| List datasets by a specific user |
| Check if a dataset exists |
Kernels / Notebooks (15 tools)
Tool | Description |
| Search and list notebooks |
| Get notebook source code |
| Push/save a notebook |
| Push local .ipynb to Kaggle |
| Download kernel output (URL) |
| Download kernel output locally |
| Check kernel execution status |
| List kernel files |
| Delete a kernel |
| Initialize kernel template locally |
| Create interactive session |
| Get session status |
| List session output files |
| Cancel running session |
| Generate kernel-metadata.json |
Models (16 tools)
Tool | Description |
| Search and list models |
| Get model details |
| Create a new model |
| Update model info |
| Delete a model |
| Get model performance metrics |
| List model instances |
| Get instance details |
| Create a new instance |
| Delete an instance |
| List instance files |
| List instance versions |
| Create new version |
| Download version files |
| List version files |
| Delete a version |
Discussions (10 tools)
Tool | Description |
| Search discussions |
| List discussions for competition/dataset |
| Get discussion content |
| Get discussion comments |
| Search across all comments |
| Browse by source type |
| Browse competition solutions |
| Browse write-ups by type |
| Browse trending discussions |
| List your discussions |
Benchmarks (1 tool)
Tool | Description |
| Get benchmark leaderboard |
Data Preview & Analysis (4 tools)
Tool | Description |
| Preview first N rows of a CSV |
| Preview any text data file |
| Analyze column types and stats |
| Diff two CSV files |
Workflow Utilities (9 tools)
Tool | Description |
| Auto-generate competition starter notebook |
| Unified search across competitions, datasets, notebooks |
| List local files with sizes |
| Monitor long-running Kaggle operations |
| Download pip wheels for offline use |
| Download requirements.txt wheels |
| Upload pip library as Kaggle dataset |
| Upload requirements as Kaggle dataset |
| Check local wheel version |
License
MIT
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP server for building and testing AI agents with multi-model experimentation and insights.
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
The CustomGPT.ai MCP server is a fully managed, RAG-powered endpoint that connects large language models with private knowledge bases and external data sources. It provides tools for retrieval-augmented generation queries (send_message), data ingestion (upload_file), and source listing, enabling AI agents to query private documents like PDFs with high accuracy and real-time citations.
Related MCP Servers
- AlicenseBqualityDmaintenanceMCP server for Kaggle API integration that allows creating, running, and managing Kaggle notebooks programmatically.81MIT
- AlicenseBqualityDmaintenanceA full-featured MCP server for the Kaggle API — competitions, datasets, kernels, models, benchmarks, and discussions.5188 PyPI4MIT
- AlicenseNot gradedqualityCmaintenanceA Model Context Protocol (MCP) server that provides seamless integration with the Kaggle API, enabling interaction with competitions, datasets, kernels, and models through MCP-compatible clients.40 PyPIMIT
- AlicenseCqualityCmaintenanceEnables AI agents to interact with Kaggle's platform for managing datasets, competitions, kernels, models, forums, and benchmarks via 68 MCP tools.682MIT