CtrlK
BlogDocsLog inGet started
Tessl Logo

testland/deepchecks-tests

Run Deepchecks suites (data integrity, train-test validation, model evaluation) on tabular / NLP / vision data + models. Pass `result.passed_conditions()` to CI to gate on regressions; the same checks run during research, CI, and production monitoring per the Deepchecks lifecycle posture. Use before training to catch train-test leakage and data-integrity defects in a tabular, NLP, or vision dataset, and to re-run the same suite on production samples to detect drift.

74

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Overview
Quality
Evals
Security
Files

Quality

Content

85%

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

The body is highly actionable and clearly sequenced with explicit gating feedback loops, and it is well-organized for a single-file skill. The only weakness is minor verbosity from the marketing quote and repeated external-reference cues that could be trimmed.

Suggestions

Drop the 'holistic open-source solution for all of your AI & ML validation needs' marketing quote from the intro; it restates capability without adding actionable guidance.

Consolidate the repeated 'Per the [Deepchecks welcome]' cues into a single References section to avoid re-pointing to the same URL in Steps 1, 2, and Limitations.

If Vision/NLP usage is common, add a short local note or a one-level-deep reference file rather than deferring entirely to the external quickstart link.

DimensionReasoningScore

Conciseness

The body is mostly lean and assumes Claude's competence (code + one-line 'Catches:' per step), but the marketing-quote opening ('holistic open-source solution for all of your AI & ML validation needs') and repeated 'Per the [Deepchecks welcome]' cues add tokens that do not earn their place.

2 / 3

Actionability

Every step ships fully executable, copy-paste-ready code with real imports, suite/check names, and a concrete CI YAML block — matching the score-3 anchor with no pseudocode gaps.

3 / 3

Workflow Clarity

A clear 7-step sequence with explicit validation checkpoints (`if not result.passed_conditions(): raise SystemExit`) in Steps 5 and 6 plus an Anti-patterns table for error recovery, matching the score-3 anchor with explicit feedback loops.

3 / 3

Progressive Disclosure

A well-organized single-file skill with clearly headed sections and one cleanly signaled external reference; with no bundle files needed, the sectioned structure satisfies the simple-skill allowance for score 3.

3 / 3

Total

11

/

12

Passed

Description

100%

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

The description is strong across all dimensions: it names concrete actions, provides explicit 'Use when' triggers for pre-training and production-drift scenarios, and occupies a distinct, low-conflict niche. Trigger terms are natural and well-covered.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — running three named suites (data integrity, train-test validation, model evaluation) and passing `result.passed_conditions()` to CI to gate on regressions — matching the score-3 anchor for multiple specific concrete actions.

3 / 3

Completeness

Explicitly answers both what (run the three suites + CI gating) and when ('Use before training...' and 're-run... on production samples to detect drift'), matching the score-3 anchor with explicit triggers rather than implied guidance.

3 / 3

Trigger Term Quality

Includes natural terms users would say — 'train-test leakage', 'data-integrity defects', 'drift', 'model evaluation' — giving good coverage rather than only technical jargon, though it leans slightly library-specific.

3 / 3

Distinctiveness Conflict Risk

Occupies a clear niche (Deepchecks ML validation) with distinct triggers (train-test leakage, drift detection) unlikely to conflict with other skills, matching the score-3 anchor.

3 / 3

Total

12

/

12

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Reviewed

Table of Contents