CtrlK
BlogDocsLog inGet started
Tessl Logo
PluginPublic

testland/deepchecks-tests

Use before training to catch train-test leakage and data integrity problems in tabular, NLP or vision data, and to rerun the same suite on production samples for drift. Runs Deepchecks data integrity, train-test validation and model evaluation suites and gates CI on passed conditions.

Quality

89%

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

81%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.

A well-structured, actionable skill with a clear sequenced workflow and explicit validation gating. The main gaps are minor: a small amount of trimmable prose, an CI step referencing a script whose contents are never shown, and no use of bundle files to split out detail.

Suggestions

Trim the marketing tagline quote and the repeated "Per the [Deepchecks welcome] page" asides — Claude only needs the link, not Deepchecks' self-description.

Make the CI step self-contained: either inline the commands that belong in ml/deepchecks_suite.py (composition of Steps 2–6) or briefly state that the script should contain the Step 2–6 code.

Consider moving the anti-patterns table and per-data-type (vision/NLP) guidance into a references/ file to keep SKILL.md a lean overview with one-level-deep pointers.

DimensionReasoningScore

Conciseness

The body is efficient — short sequenced steps, tight code blocks, and a compact anti-patterns table — but includes minor trimmable padding such as the marketing tagline quote ("a holistic open-source solution for all of your AI & ML validation needs") and repeated "Per the [Deepchecks welcome] page" asides.

4 / 5

Actionability

Every step has concrete, mostly executable code including CI YAML and per-check thresholds, but the CI step runs `python ml/deepchecks_suite.py` — a script the skill never shows how to create — and `train_df`/`model` remain undefined placeholders, so it is not fully copy-paste ready end-to-end.

4 / 5

Workflow Clarity

Steps 1–7 are clearly sequenced (install → wrap data → integrity → train-test validation → model evaluation → thresholds → CI) with explicit validation checkpoints (`if not result.passed_conditions(): raise SystemExit`), diagnostic output (`print(result.value)`), and an anti-patterns table covering common failure modes.

5 / 5

Progressive Disclosure

Sections are well-organized with a clearly signaled one-level-deep external reference, but no bundle files exist and everything (~150 lines: anti-patterns, limitations, CI config) is inlined, so content that could be split into reference files is inline.

4 / 5

Total

17

/

20

Passed

Description

92%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 that explicitly covers what the skill does and when to use it, with concrete actions and a distinctive ML-validation niche. The only weakness is slightly limited keyword variation (missing synonyms like "data validation" or "ML testing").

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — "catch train-test leakage and data integrity problems", "rerun the same suite on production samples for drift", "gates CI on passed conditions" — plus three named suites, giving comprehensive coverage in third-person voice.

5 / 5

Completeness

Both parts are explicit: the "what" ("Runs Deepchecks data integrity, train-test validation and model evaluation suites and gates CI on passed conditions") and a concrete "when" trigger clause ("Use before training... and to rerun the same suite on production samples for drift").

5 / 5

Trigger Term Quality

Natural phrases like "train-test leakage", "data integrity", "drift", and "production samples" are present, but common variations users might say ("data validation", "ML testing", "data drift", "train/test split") are missing, so coverage is good rather than comprehensive.

4 / 5

Distinctiveness Conflict Risk

It names the tool (Deepchecks) and a clear niche — ML train-test leakage, data integrity, and drift monitoring — with distinct triggers, so overlap risk with other skills is minimal.

5 / 5

Total

19

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Reviewed

On this page