Check a skill against quality criteria
Scanned 2/12/2026
Install to Claude Code
npx -y skills add wpank/ai --skill skills --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of validate-skill?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/wpank-validate-skill)More formats (shields.io, HTML) on the badges page.
---
name: validate-skill
model: fast
description: Check a skill against quality criteria
usage: /validate-skill <path>
---
# /validate-skill
Validate a skill against quality criteria.
## Usage
```
/validate-skill <path>
```
**Arguments:**
- `path` — Path to skill (can be relative or absolute)
## Examples
```
/validate-skill ai/skills/realtime/websocket-hub-patterns
/validate-skill ai/staging/skills/new-pattern
/validate-skill ai/skills/my-skill
```
## What It Does
1. **Reads** the SKILL.md file at the given path
2. **Checks** for required sections (When to Use, NEVER Do)
3. **Validates** description format (WHAT, WHEN, KEYWORDS)
4. **Measures** line count (<300 preferred, 500 max)
5. **Scans** for project-specific references
6. **Checks** for code examples
7. **Reports** pass/fail for each criterion
8. **Suggests** improvements for failures
## Sample Output
```
Validating: ai/skills/realtime/websocket-hub-patterns
Quality Criteria:
✓ Has description with WHAT, WHEN, KEYWORDS
✓ Has "When to Use" section
✓ Has code examples
✓ Has NEVER Do section
✓ Line count: 187 (< 300 ✓)
✓ Project-agnostic (no hardcoded names)
Expert Knowledge Check:
✓ Contains patterns not in base model
Result: PASS (7/7 criteria met)
```
## Quality Criteria Checked
| Criterion | Requirement |
|-----------|-------------|
| Description | Contains WHAT, WHEN, KEYWORDS |
| When to Use | Section exists with trigger conditions |
| Code Examples | At least one code block |
| NEVER Do | Section exists with anti-patterns |
| Line Count | <300 preferred, 500 max |
| Project-Agnostic | No hardcoded project names/paths |
| Expert Knowledge | >70% not in base Claude model |
## Output Locations
- No files created — displays to terminal only
## Automation
- **Optional automation:** Run `python3 scripts/validate_links.py` to validate all markdown links in the skill with fix suggestions.
## Related
- **Create skill:** `/create-skill` (guided creation)
- **Check overlaps:** `/check-overlaps` (find redundant skills)
- **Quality criteria:** [`ai/skills/extraction/references/skill-quality-criteria.md`](ai/skills/extraction/references/skill-quality-criteria.md)
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!