name: skill-audit
description: >
Audit and score OpenClaw AgentSkills against structural compliance, quality standards, and
OpenClaw-specific architecture patterns. Produces a 0-100 score with A-F grade, dimensional
breakdown, and actionable improvement recommendations. Use when asked to audit, score, validate,
check, or evaluate a skill or SKILL.md.
Keywords: skill audit, skill score, skill check, skill validate, 스킬 검증, 스킬 점수, 스킬 평가, 스킬 감사, 스킬 점검, スキル検証, スキル採点, スキル評価, スキル監査, 技能验证, 技能评分, 技能检查, 技能审计.
validate skill, skill quality, 스킬 품질.
Audit and score OpenClaw AgentSkills with a multi-dimensional scoring system.
"audit this skill" or "score path/to/SKILL.md"
Total (0–100) = Weighted Average
Dimension Weight Score Range Description
─────────────────── ─────── ─────────── ───────────
A. Structure 20% 0–100 Spec compliance, frontmatter, file structure
B. Triggering 15% 0–100 description quality, keywords, triggering
C. Style Guide 20% 0–100 Judgment criteria, preservation/compression rules
D. Workflow 15% 0–100 Phase 0, sub-agent design
E. Sub-Agent Design 15% 0–100 Context injection, templates
F. Conciseness 15% 0–100 SKILL.md length, references separation
Grade: A(90+) | B(75+) | C(60+) | D(45+) | F(<45)Frontmatter, file structure, agentskills spec compliance.
| Check | Points | Criteria |
|-------|--------|----------|
| Frontmatter exists | 15 | --- opening and closing present |
| name valid | 10 | lowercase, hyphens only, 1-64 chars |
| description present | 15 | Non-empty, 1-1024 chars |
| description has keywords | 10 | Keywords in multiple languages |
| No auxiliary files | 10 | No README.md, CHANGELOG.md, etc. |
| references/ organized | 10 | If exists, 1-level deep, TOC for >100 lines |
| SKILL.md under 500 lines | 10 | Lean body, details in references/ |
| No duplication | 5 | Info in SKILL.md OR references, not both |
| Folder name matches name | 15 | skill-name/SKILL.md |
Max: 100
Whether the description triggers the agent correctly.
| Check | Points | Criteria |
|-------|--------|----------|
| "What" described | 20 | Clear description of what it does |
| "When" in description | 25 | Trigger conditions in description (not body) |
| Keywords (Korean) | 15 | Korean keywords included |
| Keywords (English) | 15 | English keywords included |
| Not too verbose | 10 | description ≤ 300 words (metadata cost) |
| Not too vague | 15 | Not overly broad (e.g., "helpful skill") |
Max: 100
Whether judgment criteria (preservation/compression) are specified.
Applicability:
For N/A skills, exclude this dimension and redistribute weights among remaining dimensions:
| Check | Points | Criteria |
|-------|--------|----------|
| Preservation criteria | 30 | "What to preserve" with specific examples |
| Compression criteria | 25 | "What to compress" with specific examples |
| Task essence defined | 20 | Core task definition (e.g., "summarization is restructuring, not compression") |
| Style rules explicit | 15 | Specific rules for tone, style, length, etc. |
| Anti-patterns | 10 | "What not to do" specified (bonus) |
Max: 100
Whether Phase 0, execution order, and validation steps exist.
| Check | Points | Criteria |
|-------|--------|----------|
| Phase 0 (analysis) first | 30 | Full understanding step before task execution |
| Clear phase numbering | 15 | Phase 0, 1, 2... structured |
| Context injection described | 25 | Explicit statement that Phase 0 results are injected into sub-agent prompts |
| Validation step | 20 | Result quality validation step (length/style/omissions) |
| Error handling | 10 | Failure response instructions |
Max: 100
Sub-agent prompt design quality.
Applicability:
For N/A skills, exclude this dimension and redistribute weights among remaining dimensions:
| Check | Points | Criteria |
|-------|--------|----------|
| Prompt template exists | 20 | Template file in references/ |
| [ ] placeholders | 20 | Blank fields filled by Phase 0 explicitly marked |
| Context fields defined | 20 | Full context, chunk position, preservation list |
| Style rules in template | 15 | 3-4 line style instructions included |
| Merge ≠ concatenate | 15 | Explicit prohibition of simple concatenation |
| Sequential pipeline | 10 | Sequential ordering for multi-stage output |
Max: 100
Context window efficiency.
| Check | Points | Criteria |
|-------|--------|----------|
| SKILL.md body < 500 lines | 40 | Length compliance |
| SKILL.md body < 300 lines | 10 | Bonus (very concise) |
| No redundant instructions | 20 | No duplicate instructions |
| "Agent is smart" principle | 15 | Does not redundantly explain what the agent already knows |
| References linked | 15 | "When to read" description for each reference file |
Max: 100
Always do both:
~/.openclaw/workspace/skill-audit-reports/[skill-name]-audit-[YYYY-MM-DD-HHmm].md. If the same skill is audited again, a new timestamped file is created — never overwrite previous reports. Never save inside the target skill's directory — keep skill folders clean for distribution.Per the Language Policy above, generate two complete versions:
Separate them with a clear divider:
---
## English VersionIf the user requested English, a single English version suffices.
# Skill Audit: [skill-name]
**Version**: skill-audit v1.0
**Date**: YYYY-MM-DD
**Target**: path/to/SKILL.md
## Score: 78/100 (B)
| Dimension | Score | Weight | Weighted |
|-----------|-------|--------|----------|
| A. Structure | 85 | 20% | 17.0 |
| B. Triggering | 90 | 15% | 13.5 |
| C. Style Guide | 60 | 20% | 12.0 |
| D. Workflow | 70 | 15% | 10.5 |
| E. Sub-Agent Design | 80 | 15% | 12.0 |
| F. Conciseness | 85 | 15% | 12.8 |
| **Total** | | | **77.8** |
## Grade: B
---
## Dimension Details
### A. Structure (85/100)
✅ Frontmatter exists (+15)
✅ name valid: "summarize" (+10)
✅ description present (+15)
...
❌ SKILL.md over 500 lines (-10)
→ Tip: Split detailed specs into references/
### B. Triggering (90/100)
...
## Improvement Recommendations
1. **[High]** Add preservation/compression criteria to Style Guide → C score +20 expected
2. **[Medium]** Specify Phase 0 → D score +15 expected
3. **[Low]** Add references/ TOC → A score +5 expectedAudit multiple skills at once:
"Audit all skills in the skills/ folder"
→ Generate individual score cards per skill, then produce a comparison table:
| Skill | Total | Grade | A | B | C | D | E | F |
|-------|-------|-------|---|---|---|---|---|---|
| summarize | 82 | B+ | 93 | 88 | 88 | 77 | 78 | 65 |
| changelog | 85 | B+ | 90 | 85 | 90 | 85 | N/A | 80 |
| scaffold | 81 | B | 85 | 80 | 75 | 80 | N/A | 85 |references/scoring-rubric.md — Detailed scoring rubric with examples per dimensionexamples/ — Completed audit reports and batch comparisons for validated skills