Tests and scores any Agent Skill against the official anthropics/skills specification. Use this skill when you need to check if a skill repository or SKILL.
概览
An Agent Skill that tests and scores any other Agent Skill against the official anthropics/skills specification. 🦞 clawhub here: https://clawhub.ai/youngfreefjs/skill-test-skill Install this skill, then point it at any skill directory or SKILL.md file. It will: 1. Read every file in the skill (SKILL.md, scripts/, references/, assets/, etc.) 2. Check each file line-by-line against the official spec rules 3. Score the skill across (100 points total) 4. Output a detailed Markdown report with findings and prioritized improvement suggestions Supports output — the report language follows the language of your request. After installing the skill, you can trigger it with: - "Check my skill at /path/to/my-skill" - "Score this skill: https://github.com/user/repo/tree/main/skills/my-skill" - "Does my SKILL.md follow the spec?" - "检查一下 /path/to/my-skill 是否符合规范" - "给这个skill打分：/path/to/my-skill" - "这个skill符合anthropics的要求吗？" The validation rules in this skill are based on:
README
以下为原文 README 节选，完整内容请查看原文。
skill-test-skill
An Agent Skill that tests and scores any other Agent Skill against the official anthropics/skills specification.
🦞 clawhub here: https://clawhub.ai/youngfreefjs/skill-test-skill
What It Does
Install this skill, then point it at any skill directory or SKILL.md file. It will:
- Read every file in the skill (SKILL.md, scripts/, references/, assets/, etc.)
- Check each file line-by-line against the official spec rules
- Score the skill across 6 dimensions (100 points total)
- Output a detailed Markdown report with findings and prioritized improvement suggestions
Supports Chinese and English output — the report language follows the language of your request.
Scoring Dimensions
| # | Dimension | Max Points |
|---|---|---|
| 1 | Directory Structure | 10 |
| 2 | Frontmatter Compliance (name, description, etc.) | 30 |
| 3 | Body Content Quality | 25 |
| 4 | Progressive Disclosure Design | 15 |
| 5 | Optional Directory Quality (scripts/, references/, assets/) | 10 |
| 6 | Description Trigger Optimization | 10 |
Grade Scale
| Score | Grade |
|---|---|
| 90–100 | Excellent — production-ready |
| 75–89 | Good — minor improvements recommended |
| 60–74 | Acceptable — needs improvement |
| 40–59 | Poor — significant rework required |
| 0–39 | Critical — major rewrite needed |
Usage Examples
After installing the skill, you can trigger it with:
English:
"Check my skill at /path/to/my-skill""Score this skill: https://github.com/user/repo/tree/main/skills/my-skill""Does my SKILL.md follow the spec?"
中文：
"检查一下 /path/to/my-skill 是否符合规范""给这个skill打分：/path/to/my-skill""这个skill符合anthropics的要求吗？"
Rules & Specification
The validation rules in this skill are based on:
- Agent Skills Specification — the official spec defining SKILL.md format, frontmatter fields, directory structure, and progressive disclosure requirements
- anthropics/skills — skill-creator — Anthropic’s official skill-creation guide, which defines best practices for writing style, description optimization, trigger tuning, and the “why over MUST” principle
- anthropics/skills repository — real-world skill examples used as reference for what good looks like
Key principles derived from skill-creator:
- All “when to use” information belongs in the
descriptionfield, not the body - Prefer explaining why over issuing bare
ALWAYS/NEVERcommands - Keep
SKILL.mdunder 500 lines; move large content toreferences/ - The
descriptionfield is the primary triggering mechanism — make it specific and slightly “pushy” to prevent undertriggering
File Structure
skill-test-skill/
├── SKILL.md ← Core skill instructions (416 lines)推荐工具
换一个关键词，或者移除筛选条件。
安装
npx skillfish add youngfreefjs/skill-test-skill