A
licenseNot graded
qualityF
maintenanceEnables AI agents and CLI users to drive Kaggle notebooks end-to-end—pull, edit, save new versions, run, monitor, and fetch logs—using a local stdio MCP server with a Kaggle API token.
MIT
Kaggle MCP Server
by art-qalam-fr
Related Servers
Alternatives to Kaggle MCP Server
No user-submitted related servers found.
Related Servers
- AlicenseBqualityAmaintenanceA local stdio MCP server that gives Claude on-demand access to any Kaggle competition's brief, rules, evaluation, data description, leaderboard, top discussions and public notebooks, using the user's own Kaggle credentials and a disk cache with no hosted infrastructure. It also exposes a live tool for spotting forum topics and comments posted since the last fetch.102MIT
- AlicenseBqualityDmaintenanceA full-featured MCP server for the Kaggle API — competitions, datasets, kernels, models, benchmarks, and discussions.5186 PyPI4MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI assistants to interact with Kaggle competitions, including listing competitions, downloading files, submitting predictions, and viewing submission history.10MIT
- AlicenseNot gradedqualityDmaintenanceA full-featured MCP server with 96 tools for the Kaggle API, enabling users to manage competitions, datasets, notebooks, models, discussions, and workflows via natural language.MIT
- FlicenseNot gradedqualityCmaintenanceEnables AI agents to interact with Kaggle through MCP, including remote GPU/TPU kernel execution, competition and dataset downloads, and kernel management via SSE.-
TDQS
C2.9/5.0
Scored across 51 tools
Disambiguation2/5
Multiple tools have overlapping purposes: discussions_trending, discussions_search, discussions_list, discussions_by_source, discussions_solutions, and discussions_writeups all retrieve discussion content with subtle differences; competition_leaderboard vs competition_leaderboard_download also confuse. Kernel_output's description mismatches its name, further muddying tool selection.
Naming Consistency4/5
Names mostly follow snake_case with domain prefixes (competition_, dataset_, model_), but plural/singular inconsistency exists (discussions_* vs discussion_*; competitions_list vs competition_leaderboard). Some tool names are less predictable (kernel_output, file_upload), though the overall pattern remains readable.
Tool Count1/5
51 tools is extreme over-provisioning for an MCP server and exceeds the 50+ threshold for severe mismatch. Even for Kaggle's broad surface, many tools could be consolidated (e.g., six discussion-listing variants), making the set unwieldy.
Completeness4/5
The server covers CRUD for datasets and models, plus competition submissions, kernel sessions, and discussion reading. Minor gaps exist (no update/delete for model instances, no discussion posting/reply), but core workflows are supported.
Maintenance
ActivityMaintained
ResponsivenessNo issues