skill-reviewer
Reviews and improves Claude Code skills against official best practices. Supports three modes - self-review (validate your own skills), external review (evaluate others' skills), and auto-PR (fork, improve, submit). Use when checking skill quality, reviewing skill repositories, or contributing improvements to open-source skills.
76
63%
Does it follow best practices?
Impact
100%
1.33xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./skill-reviewer/SKILL.mdQuality
Discovery
85%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong description that clearly communicates what the skill does (three review modes for Claude Code skills) and when to use it. The three-mode breakdown adds helpful specificity. The main weakness is that trigger terms could be broader to capture more natural user phrasings.
Suggestions
Add more natural trigger terms users might say, such as 'SKILL.md', 'lint skill', 'validate skill file', or 'skill template' to improve discoverability.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple concrete actions: 'Reviews and improves Claude Code skills', and describes three specific modes — self-review (validate your own skills), external review (evaluate others' skills), and auto-PR (fork, improve, submit). These are concrete, actionable capabilities. | 3 / 3 |
Completeness | Clearly answers both 'what' (reviews and improves skills against best practices with three modes) and 'when' (explicit 'Use when checking skill quality, reviewing skill repositories, or contributing improvements to open-source skills'). | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'skill quality', 'skill repositories', 'open-source skills', 'review', and 'best practices', but misses natural variations users might say such as 'lint skills', 'SKILL.md', 'skill file', 'validate skill', or 'skill template'. The terms are somewhat niche and may not cover all natural phrasings. | 2 / 3 |
Distinctiveness Conflict Risk | This is a very specific niche — reviewing Claude Code skills against best practices — which is unlikely to conflict with other skills. The three modes (self-review, external review, auto-PR) further distinguish it. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill covers a useful meta-task (reviewing other skills) with a clear three-mode structure and helpful concrete examples for common issues and PR etiquette. However, it suffers significantly from referencing multiple supporting files that don't exist in the bundle, undermining both actionability and progressive disclosure. The workflows would benefit from explicit validation steps and error handling rather than simple checkbox lists.
Suggestions
Include the referenced bundle files (evaluation_checklist.md, pr_template.md, marketplace_template.json) or inline their essential content to make the skill self-contained and actionable.
Add explicit validation and error recovery steps to the External Review and Auto-PR workflows (e.g., what to do if cloning fails, how to verify improvements before PR submission).
Remove or condense the setup section — the auto-install sequence references a specific plugin ecosystem that may not be accessible, and the find commands are fragile. Consider making the skill work without this dependency.
Tighten the PR Guidelines section by combining the tone examples and required sections into a more compact format, or move them entirely to the referenced pr_template.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is moderately efficient but includes some unnecessary structure like the setup section with bash commands for installing a dependency that may not exist, and the checklist table format adds bulk. Some sections like PR guidelines and tone examples, while useful, could be tighter. | 2 / 3 |
Actionability | Provides concrete bash commands and YAML examples for common fixes, but several key workflows rely on references to files that don't exist in the bundle (e.g., references/evaluation_checklist.md, references/pr_template.md). The setup commands reference a specific plugin ecosystem that may not be real or accessible, reducing practical executability. | 2 / 3 |
Workflow Clarity | Three modes are clearly identified with checklists, but the workflows lack explicit validation checkpoints and error recovery steps. The External Review and Auto-PR workflows are presented as simple checkbox lists without concrete commands or feedback loops (e.g., what to do if the fork fails, if validation finds issues, etc.). | 2 / 3 |
Progressive Disclosure | The skill references four bundle files (references/evaluation_checklist.md, references/pr_template.md, references/marketplace_template.json, and an external URL) but none of these files exist in the bundle. This creates broken references and a hollow progressive disclosure structure that promises depth but delivers nothing. | 1 / 3 |
Total | 7 / 12 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 — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- daymade/claude-code-skills
- Commit
b2003d6
- 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.