explaining-machine-learning-models

Explain trained machine learning models through feature attribution, local explanations, and behavior summaries. Use as an explicit/manual helper once a model already exists, not for training ownership, leakage auditing, or general ML strategy selection.

Quality

57%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./bundled/skills/explaining-machine-learning-models/SKILL.md

Quality

Content

43%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 is a tidy overview but lacks any executable guidance or sequenced workflow, and it fails to link to the bundled scripts and assets that would make it actionable. The 'When to Use' bullets are phrased as goals ('Understand why...') rather than trigger phrases, blurring when the skill applies.

Suggestions

Add a concrete quick-start example showing how to invoke explain_model.py or feature_importance.py with model and data inputs, so the skill is copy-paste actionable.

Turn 'Typical Outputs' into a short ordered workflow (load model -> choose SHAP/LIME -> generate explanation -> surface caveats) with validation of explanation stability.

Reference the bundled scripts/assets explicitly from the body (e.g. 'See scripts/explain_model.py') and remove or create the missing references/*.md files listed in references/README.md.

Dimension	Reasoning	Score
Conciseness	The body is lean and avoids explaining concepts Claude already knows, with only minor instances (e.g. the Positioning section restates the description) that could be trimmed.	4 / 5
Actionability	The body only describes outputs ('Feature importance or attribution summaries', 'Local explanation workflow') with no executable code, commands, or concrete steps, despite bundled scripts like explain_model.py existing unused.	2 / 5
Workflow Clarity	There is no sequenced multi-step process; 'Typical Outputs' is a loose list of artifacts rather than an ordered workflow, and no validation checkpoints exist for explanation workflows.	2 / 5
Progressive Disclosure	Section structure is present and clean, but the body never references the bundled scripts/assets (orphaned files) and the references/ directory only lists .md files that do not actually exist, so navigation to detail is not effectively signaled.	3 / 5
	Total	11 / 20 Passed

Description

71%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.

The description is well-scoped with clear capability actions and strong boundary guidance that distinguishes it from related skills. Its main weakness is trigger-term quality: the language leans technical and omits the natural phrases a user would say when reaching for interpretability help.

Suggestions

Add natural user-facing trigger phrases, e.g. 'Use when the user asks why a model made a specific prediction, what features drove an output, or how to explain a model to stakeholders'.

Include common synonyms and library-adjacent terms users say ('model interpretability', 'feature importance', 'SHAP/LIME') to improve trigger matching.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('feature attribution, local explanations, and behavior summaries') but stops short of the comprehensive multi-action coverage of the 5-anchor example.	4 / 5
Completeness	Clearly states what it does and gives explicit 'when' guidance ('Use as an explicit/manual helper once a model already exists'), but the trigger framing is about positioning rather than concrete user-utterance triggers.	4 / 5
Trigger Term Quality	Includes relevant terms like 'feature attribution' and 'local explanations' but these are technical; it misses natural layman phrasings users would actually say such as 'why did the model predict X' or 'model interpretability'.	3 / 5
Distinctiveness Conflict Risk	Carves a clear niche (post-training interpretability) with explicit exclusions for training, leakage auditing, and ML strategy, minimizing overlap with sibling skills.	5 / 5
	Total	16 / 20 Passed

Validation

87%

Warnings & errors only

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

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: foryourhealth111-pixel/Vibe-Skills
Path: bundled/skills/explaining-machine-learning-models/SKILL.md
Commit: d5ae560

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

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.