agentleFS
jara-r-kjara-r-k/claude-code-skills/skills/skill-compliance-checklist/SKILL.md

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.

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
  1. Very specific:

    File naming: Must be exactly SKILL.md (case-sensitive) inside a kebab-case folder

    line 30
  2. Very specific:

    YAML frontmatter: Must have --- delimiters, name (kebab-case), description (WHAT + WHEN)

    line 31
  3. Very specific:

    No XML angle brackets (< or >)

    line 36
  4. Distinctive:

    Add negative triggers: "Do NOT use for [unrelated topic]"

    line 89

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

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:

  1. File naming: Must be exactly SKILL.md (case-sensitive) inside a kebab-case folder
  2. YAML frontmatter: Must have --- delimiters, name (kebab-case), description (WHAT + WHEN)
  3. 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
  4. Body structure: Clear headings, actionable instructions, not vague
  5. Progressive disclosure: Large reference content in references/, not inline
  6. Error handling: Documents what to do when things fail
  7. Examples: Concrete usage examples provided
  8. 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 ## Important or ## Critical headers
  • 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 ## Critical headers
  • Ambiguous language — replace "validate properly" with specific checks

Examples

Auditing a single skill:

/skill-compliance-checklist ~/.claude/skills/persona-audit/SKILL.md

Auditing all skills:

/skill-compliance-checklist --all

Quick check after creating a new skill:

/skill-compliance-checklist ./my-new-skill/SKILL.md

Download now (ZIP)

Does 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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.