Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Audit Skills

ASecurity

Audits skills, commands, and prompt templates for best practice violations including length, checklists, verification steps, and progressive disclosure. Use after creating or modifying skills, during reviews, or to improve existing prompt files. Produces prioritized improvement suggestions.

39 stars
0 votes
0 copies
0 views
Added 9/20/2026
code-quality

Security Analysis

A100/100

Scanned 9/20/2026

Install to Claude Code

$npx -y skills add benjaminshoemaker/ai_coding_project_base --skill audit-skills --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Audit Skills?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Audit Skills
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/benjaminshoemaker-audit-skills/badge)](https://www.skillsdirectory.com/skills/benjaminshoemaker-audit-skills)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: audit-skills
description: Audits skills, commands, and prompt templates for best practice violations including length, checklists, verification steps, and progressive disclosure. Use after creating or modifying skills, during reviews, or to improve existing prompt files. Produces prioritized improvement suggestions.
---

# Audit Skills

Analyze skills, legacy commands, and prompt templates in this repository against best practices and produce prioritized improvements.

## Workflow

Copy this checklist and track progress:

```
Audit Progress:
- [ ] Step 1: Discover all auditable files (skills, commands, prompts)
- [ ] Step 2: Load criteria
- [ ] Step 3: Audit each file
- [ ] Step 4: Score and prioritize
- [ ] Step 5: Output report
```

### Step 1: Discover All Auditable Files

Find all skill-like files in the repository:

Use Glob to find files matching these patterns:
- `.claude/skills/*/SKILL.md` — Skills
- `.claude/commands/*.md` — Legacy commands
- `.claude/skills/*/PROMPT.md` — Co-located prompts
- `*_PROMPT.md` — Root-level prompt templates

Classify each file:

| Type | Example | Criteria Applied |
|------|---------|-----------------|
| **Skill** | `.claude/skills/*/SKILL.md` | All criteria (L, C, S, E, D) |
| **Command** | `.claude/commands/*.md` | All criteria (L, C, S, E, D) — same frontmatter as skills |
| **Prompt** | `PROMPT.md`, `*_PROMPT.md` | L, C, S, E only — no frontmatter, so skip D1-D3 |

Record the total count across all types for verification later.

### Step 2: Load Criteria

Read [CRITERIA.md](CRITERIA.md) for the full audit checklist.

Read [SCORING.md](SCORING.md) for severity definitions.

### Step 3: Audit Each File

For each discovered file:

1. **Read the file** — note line count immediately
2. **Check applicable criteria** from CRITERIA.md (see type table in Step 1)
3. **For skills with co-located PROMPT.md** — audit the PROMPT.md too and report alongside the SKILL.md (they function as a unit)
4. **Cross-reference check** — Grep for `/skill-name` invocation patterns in other SKILL.md files. If explicitly invoked (not merely mentioned in prose), apply CO1 (output contract) check
5. **Record violations** with criterion ID and evidence

Use this template per file:

```
File: <path>
Type: Skill | Command | Prompt
Lines: <count> (+ <N> in PROMPT.md if applicable)
Violations:
- <ID>: <brief evidence>
```

### Step 4: Score and Prioritize

Group findings by severity (from SCORING.md):
- **Critical**: Likely to cause execution failures
- **Medium**: Degrades quality but may still work
- **Low**: Polish items

Within each severity, sort by:
1. Most violations in single skill (fix skill holistically)
2. Most common violation across skills (systemic issue)

**Systemic pattern detection**: If any single criterion appears in >50% of audited skills *and* the sample size is 5 or more, flag it as a systemic issue requiring a template-level fix rather than per-skill remediation.

### Step 5: Output Report

Use this format:

```markdown
# Skill Audit Report

Generated: <date>
Skills audited: <N>
Total findings: <N>

## Critical (fix first)

- **<skill-name>**: <violation summary> (<criterion-id>)
  - Evidence: <quote or line numbers>
  - Suggested fix: <concrete action>

## Medium Priority

...

## Low Priority

...

## Top 3 Recommendations

1. **<most impactful fix>** — affects <N> skills (<criterion-ids>)
2. **<second most impactful>** — affects <N> skills (<criterion-ids>)
3. **<third most impactful>** — affects <N> skills (<criterion-ids>)

{If any criterion appears in >50% of skills}
## Systemic Issues

- **<criterion-id>**: Found in <N>/<total> skills — consider a template-level fix
{/If}

## Summary by Criterion

| ID | Description | Count |
|----|-------------|-------|
| C1 | Missing checklist | 5 |
| ... | ... | ... |
```

## Error Handling

| Situation | Action |
|-----------|--------|
| CRITERIA.md or SCORING.md not found | Report the missing file and stop — audit cannot proceed without criteria definitions. |
| No auditable files found (no skills, commands, or prompts) | Report that no files were discovered and stop. |
| Individual file unreadable | Log the file as "skipped (unreadable)" in the report and continue with remaining files. |

---

**REMINDER**: After completing the audit, verify your report includes all discovered files (skills, commands, and prompts). Missing files indicates incomplete analysis.

Attribution

benjaminshoemakerbenjaminshoemaker
View sourceMore from benjaminshoemaker →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Caveman Review

Ultra-compressed code review comments. Cuts noise from PR feedback while preserving the actionable signal. Each comment is one line: location, problem, fix. Use when user says "review this PR", "code review", "review the diff", "/review", or invokes /caveman-review. Auto-triggers when reviewing pull requests.

1023331 votes

Caveman Commit

Ultra-compressed commit message generator. Cuts noise from commit messages while preserving intent and reasoning. Conventional Commits format. Subject ≤50 chars, body only when "why" isn't obvious. Use when user says "write a commit", "commit message", "generate commit", "/commit", or invokes /caveman-commit. Auto-triggers when staging changes.

1023331 votes

Springboot Verification

Verification loop for Spring Boot projects: build, static analysis, tests with coverage, security scans, and diff review before release or PR.

2456590 votes

Verification Loop

一个全面的 Claude Code 会话验证系统。

2456590 votes

Django Verification

Verification loop for Django projects: migrations, linting, tests with coverage, security scans, and deployment readiness checks before release or PR.

2456590 votes
View all in code-quality →