skill-quality-review

Review a SKILL.md file for adherence to Anthropic's official agent-skill best practices. Use when the user asks to "review a skill", "check skill quality", "audit skill", "validate skill quality", or before merging a PR that adds or modifies a skill in skills/. Complements the deterministic structural validator (validate_skills.py) by judging content quality — concision, description discoverability, naming, freedom calibration, progressive disclosure.

Quality

86%

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

Quality

Content

78%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, highly actionable instruction-only review skill: concrete commands, an exact output template, and proper progressive disclosure of the large reference doc. The main weakness is redundancy — the deterministic-validator boundary is restated several times, which slightly inflates the body.

Suggestions

Consolidate the 'complements / does not duplicate the deterministic validator' point: state it once in the intro and once in the Reference section, then remove the restated scope reminder at line 83 to cut ~3 lines of repetition.

Promote the Anti-Patterns grounding rules ('Cite literally', 'Do not invent rules') into an explicit final workflow step — e.g. 'Before emitting findings, verify each has a real quote and a real reference-doc section' — to give the workflow a visible quality-gate checkpoint.

Trim the per-dimension preamble that re-explains what the deterministic layer already covers (e.g. the inline 'Determinístic check can't...' clauses) to a single short parenthetical, since the boundary is established upstream.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and well-organized, but the 'complements / does not duplicate the deterministic validator' boundary is restated ~4 times (intro line 11, the scope reminder at line 83, and the reference entries at 137-138), which the skill's own concision dimension would flag as redundant restatement.	3 / 5
Actionability	Provides copy-paste-ready concrete guidance: an exact `git diff --name-only <base>... -- 'skills/**/SKILL.md'` command, per-dimension PASS/WARN/FAIL criteria with example quotes, and a complete output-format template to fill — fully actionable for an instruction-only skill.	5 / 5
Workflow Clarity	The 'How to Run' section gives a clear 6-step numbered sequence with a per-dimension PASS/WARN/FAIL decision gate and a final verdict (Approve / Approve with comments / Request changes); the grounding check exists only implicitly in the Anti-Patterns section rather than as an explicit workflow checkpoint.	4 / 5
Progressive Disclosure	The ~1100-line Anthropic best-practices doc is correctly split into references/anthropic-best-practices.md (a real bundle file) loaded on demand, while the SKILL.md body stays an overview with a clearly signaled one-level-deep reference — matching the score-5 anchor.	5 / 5
	Total	17 / 20 Passed

Description

95%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: third-person, explicit 'Use when...' clause with natural trigger phrases, clear what/when, and a well-delineated niche that distinguishes it from the structural validator. Only minor room to tighten the action list toward the score-5 specificity anchor.

Dimension	Reasoning	Score
Specificity	Lists several specific review actions/quality dimensions — 'concision, description discoverability, naming, freedom calibration, progressive disclosure' — naming the domain and multiple concrete judging tasks; not quite the comprehensive single-verb action list of a 5.	4 / 5
Completeness	Explicitly answers both what ('Review a SKILL.md file for adherence to Anthropic's official agent-skill best practices') and when via a 'Use when...' clause with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes four natural user phrasings ('review a skill', 'check skill quality', 'audit skill', 'validate skill quality') plus the SKILL.md/skills/ file references — comprehensive coverage of terms a user would actually say.	5 / 5
Distinctiveness Conflict Risk	Clear niche (skill quality/best-practice review) with distinct triggers and an explicit boundary against the deterministic validator, giving minimal conflict risk with other skills.	5 / 5
	Total	19 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
referenced_paths_exist	Referenced path issues: 4 missing, 3 deeper-than-1-level	Warning

	Total	15 / 16 Passed

Repository: mulesoft/mulesoft-dx
Path: .claude/skills/skill-quality-review/SKILL.md
Commit: 4cf0cf6

Reviewed: 5 days 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.