CtrlK
BlogDocsLog inGet started
Tessl Logo

kaggle-learner

This skill should be used when the user asks to "learn from Kaggle", "study Kaggle solutions", "analyze Kaggle competitions", or mentions Kaggle competition URLs. Provides access to extracted knowledge from winning Kaggle solutions across NLP, CV, time series, tabular, and multimodal domains.

56

Quality

66%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/kaggle-learner/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

42%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body has a sensible category-based structure pointing to a real reference tree, but is padded with duplicated listings, conceptual overview prose, and a triple-defined key term. It offers no executable commands or validation steps, and two of the five advertised knowledge directories are broken.

Suggestions

Remove the duplicated directory listing in Additional Resources (it repeats the Knowledge Categories table) and consolidate the three separate definitions of 前排方案详细技术分析 into the single Knowledge Extraction Standard section.

Fix the broken references: create `references/knowledge/cv/` and `references/knowledge/multimodal/` (their files are currently only in `.archive/`) or update the table to the directories that actually exist.

Make the learning workflow concrete: specify how the kaggle-miner agent is invoked for a given competition URL and add a verification step that the resulting markdown file appears in the correct domain directory.

DimensionReasoningScore

Conciseness

Several padded/duplicated sections: the Overview explains background Claude already knows ("Kaggle competitions are at the forefront of practical machine learning. Winning solutions often innovate..."), the Additional Resources section repeats the same directory list already given in the Knowledge Categories table, and "前排方案详细技术分析" is defined/repeated three times (Quick Reference, checklist, and its own format section). This matches anchor 2 ('noticeably verbose; several unnecessary explanations or padded sections') rather than 3, where only isolated tightening would be needed.

2 / 5

Actionability

Some concrete guidance exists: exact directory paths per domain, example filenames, and a concrete markdown template for the extraction format. However, there are no commands or executable steps — "The kaggle-miner agent will extract the winning solution" is a high-level hint with no invocation instruction, and no script exists in the bundle. This sits between anchors 2 (high-level hints, missing specific steps) and 4 (mostly executable guidance), matching 3.

3 / 5

Workflow Clarity

The "To learn from a competition" workflow lists 4 steps in sequence, and browsing guidance points to concrete paths. But there are no validation checkpoints (e.g. verifying the new knowledge file landed in the right domain directory) and the critical step 2 is delegated to an undefined agent with no instructions. This matches anchor 3 ('steps listed but validation gaps; checkpoints missing or implicit').

3 / 5

Progressive Disclosure

The structure is good in form — a category table mapping domains to `references/knowledge/[domain]/` directories, one level deep, clearly signaled — but two of the five referenced directories (`references/knowledge/cv/` and `references/knowledge/multimodal/`) do not exist (their files sit under `.archive/`), and directory content is duplicated inline in Additional Resources. Broken references plus inline duplication match anchor 3 ('some structure but could be better organized') rather than 4 ('references mostly clear, minor gaps').

3 / 5

Total

11

/

20

Passed

Description

90%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description with explicit, natural trigger phrases and a clear what/when structure in the Kaggle niche. The only weakness is that the capability statement is a single generic action ("provides access to extracted knowledge") rather than an enumeration of concrete things the skill does.

Suggestions

Replace "Provides access to extracted knowledge" with 2-3 concrete capabilities, e.g. "Extracts winning techniques, code templates, and best practices from top Kaggle solutions".

Name the concrete outputs users get, such as per-competition writeups with top-20 solution breakdowns, so the 'what' is as specific as the 'when'.

DimensionReasoningScore

Specificity

"Provides access to extracted knowledge from winning Kaggle solutions" names the domain and one main action, but the capability verbs ("learn", "study", "analyze") serve mostly as triggers rather than a list of concrete capabilities like extracting, summarizing, or applying techniques. This matches anchor 3 ('names domain and 1-2 concrete actions, but not comprehensive') rather than 4, which expects several distinct specific actions.

3 / 5

Completeness

It explicitly answers 'when' ("This skill should be used when the user asks to... or mentions Kaggle competition URLs") with concrete trigger phrases, and 'what' ("Provides access to extracted knowledge from winning Kaggle solutions across NLP, CV, time series, tabular, and multimodal domains"). Both halves are present and concrete, matching the anchor 5 example structure exactly.

5 / 5

Trigger Term Quality

"learn from Kaggle", "study Kaggle solutions", "analyze Kaggle competitions", "Kaggle competition URLs" are natural phrases a user would say, covering the common variations of the request. This matches the comprehensive anchor 5; it is not score 4 because no commonly-used natural phrasing for this domain is obviously missing.

5 / 5

Distinctiveness Conflict Risk

"Kaggle" and "Kaggle competition URLs" form a clear niche with distinct triggers; no other skill would plausibly fire on these phrases. Minimal conflict risk matches anchor 5; the domain anchor is highly specific rather than overlapping.

5 / 5

Total

18

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

referenced_paths_exist

Referenced path issues: 4 missing, 10 deeper-than-1-level

Warning

Total

14

/

16

Passed

Repository
Galaxy-Dawn/claude-scholar
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.