skill-compliance-checklist
Skill1 starsChanged 47 days agoMITView on GitHub
Audit and validate Claude Code skills against Anthropic's official guide. Use when creating new skills, reviewing existing skills, running a compliance check, asked to 'audit skills', 'check skill quality', or 'validate SKILL.md'. Covers structure, frontmatter, descriptions, progressive disclosure, error handling, and testing. Do NOT use for writing skills from scratch, deploying applications, or general code reviews.
In this repository16 other files it gives its agents.
- agents/AGENTS.mdAGENTS.md
- AGENTS.mdAGENTS.md
- examples/AGENTS.mdAGENTS.md
- .github/AGENTS.mdAGENTS.md
- .github/workflows/AGENTS.mdAGENTS.md
- scripts/AGENTS.mdAGENTS.md
- skills/AGENTS.mdAGENTS.md
- skills/figma-handoff/AGENTS.mdAGENTS.md
- skills/github-pr-review/AGENTS.mdAGENTS.md
- skills/gmail-workflow/AGENTS.mdAGENTS.md
- skills/skill-compliance-checklist/AGENTS.mdAGENTS.md
- skills/skill-compliance-checklist/references/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- figma-handoffSkill
- github-pr-reviewSkill
- gmail-workflowSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 30
File naming: Must be exactly SKILL.md (case-sensitive) inside a kebab-case folder
- Very specific: line 31
YAML frontmatter: Must have --- delimiters, name (kebab-case), description (WHAT + WHEN)
- Very specific: line 36
No XML angle brackets (< or >)
- Distinctive: line 89
Add negative triggers: "Do NOT use for [unrelated topic]"
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Skill Compliance Checklist
Validate any Claude Code skill against Anthropic's official guide. Run this before uploading, after major changes, or during periodic audits.
Instructions
Step 1: Identify Skills to Audit
Scan these locations for SKILL.md files:
~/.claude/skills/*/SKILL.md— global skills~/.claude/scheduled-tasks/*/SKILL.md— scheduled task skills.claude/agents/*.md— project-level agent skills- Any path the user specifies
Step 2: Structural Check
For each skill, verify:
- File naming: Must be exactly
SKILL.md(case-sensitive) inside a kebab-case folder - YAML frontmatter: Must have
---delimiters,name(kebab-case),description(WHAT + WHEN) - Description quality:
- Includes what the skill does
- Includes when to use it (trigger phrases users would say)
- Under 1024 characters
- No XML angle brackets (
<or>) - No "claude" or "anthropic" in the name
- Body structure: Clear headings, actionable instructions, not vague
- Progressive disclosure: Large reference content in
references/, not inline - Error handling: Documents what to do when things fail
- Examples: Concrete usage examples provided
- Word count: SKILL.md body under 5,000 words
Step 3: Content Quality Check
- Instructions are specific and actionable (not "validate things properly")
- Critical instructions use
## Importantor## Criticalheaders - Hard-coded IDs/paths are externalised to CLAUDE.md or references
- Bundled resources are clearly referenced with file paths
Step 4: Generate Report
For each skill, produce a compliance table:
| Criterion | Status | Notes |
|-------------------------|--------|-------|
| Kebab-case folder name | | |
| SKILL.md exact name | | |
| YAML --- delimiters | | |
| Name field kebab-case | | |
| Description WHAT+WHEN | | |
| Description < 1024 char | | |
| No XML in description | | |
| Progressive disclosure | | |
| Error handling section | | |
| Examples section | | |
| Body < 5,000 words | | |
| No hard-coded IDs | | |Step 5: Recommend Fixes
For any FAIL items, provide specific guidance referencing references/compliance-template.md and references/guide-patterns.md.
Troubleshooting
Skill won't upload
- Error: "Could not find SKILL.md" — rename file to exactly
SKILL.md - Error: "Invalid frontmatter" — check YAML
---delimiters and formatting - Error: "Invalid skill name" — use kebab-case only, no spaces or capitals
Skill doesn't trigger
- Description too generic (e.g., "Helps with projects")
- Missing trigger phrases users would actually say
- Debug: Ask Claude "When would you use the [skill name] skill?"
Skill triggers too often
- Add negative triggers: "Do NOT use for [unrelated topic]"
- Be more specific in description scope
- Clarify scope boundaries
Instructions not followed
- Instructions too verbose — move detail to
references/ - Critical steps buried — use
## Criticalheaders - Ambiguous language — replace "validate properly" with specific checks
Examples
Auditing a single skill:
/skill-compliance-checklist ~/.claude/skills/persona-audit/SKILL.mdAuditing all skills:
/skill-compliance-checklist --allQuick check after creating a new skill:
/skill-compliance-checklist ./my-new-skill/SKILL.mdDoes it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

