kaggle-context
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_competitionsA | List competitions already fetched on this machine (the default is marked). |
| fetch_competitionB | Fetch (or refresh) a competition from Kaggle by URL or slug, then return its brief. |
| get_briefC | Key facts (metric, deadlines, limits, submission format, external-data rule) and what else is available. |
| get_sectionC | Full text of one section: overview, evaluation, rules, data, leaderboard, timeline or prizes. |
| list_discussionsC | Saved discussion topics (solution write-ups first) with ids for get_discussion. |
| search_discussionsB | Keyword search over saved topics and comments, plus titles of other topics. |
| get_discussionC | One topic with its opening post and top comments. Fetches live if not saved. |
| list_top_notebooksC | Top public notebooks by votes, with refs for get_notebook. |
| get_notebookB | A notebook's code and markdown (outputs stripped), with attribution. ref = owner/notebook. |
| get_whats_newC | Live: topics created or commented on since |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| start-competition | Load a Kaggle competition's key facts and plan the work. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 10 tools
Most tools target distinct resources or actions, but fetch_competition says it returns the brief while get_brief also returns key facts, creating minor overlap in when to use each. Discussion and notebook tools are clearly separated.
All tools use snake_case with predictable verb_noun patterns (list_*, fetch_*, get_*, search_*), and the one compound list_top_notebooks remains readable and consistent.
Ten tools is well-scoped for a Kaggle context server, covering competitions, discussions, notebooks, and updates without excessive surface area.
Core lifecycle for fetching and reading competition context is covered, including sections, discussions, notebooks, and live updates. Minor gaps exist around general competition search or non-top notebook discovery, but agents can work around them via fetch/reference tools.