Comprehensive quality auditing and evaluation against the highest industry standards
What This Skill Does
The quality-auditor skill evaluates tools, frameworks, systems, and codebases across 12 critical dimensions with rigorous, evidence-based scoring.
12 Evaluation Dimensions
- Code Quality - Structure, patterns, maintainability
- Architecture - Design, scalability, modularity
- Documentation - Completeness, clarity, accuracy
- Usability - User experience, learning curve
- Performance - Speed, efficiency, optimization
- Security - Vulnerabilities, best practices
- Testing - Coverage, quality, automation
- Maintainability - Technical debt, refactorability
- Developer Experience - Ease of use, tooling
- Accessibility - ADHD-friendly, inclusive design
- CI/CD - Automation, deployment
- Innovation - Novelty, forward-thinking
When to Use
Use this skill when you need to:
- ✅ Audit your own tools - Measure quality objectively
- ✅ Evaluate third-party tools - Make informed decisions
- ✅ Conduct code reviews - Comprehensive assessment
- ✅ Benchmark against competitors - See how you stack up
- ✅ Identify improvements - Get actionable recommendations
- ✅ Validate quality claims - Evidence-based verification
- ✅ Prepare for production - Pre-launch quality gate
- ✅ Due diligence - Investment or acquisition evaluation
How It Works
3-Phase Process
Phase 1: Discovery (10 min)
- Read all documentation
- Examine codebase
- Test functionality
- Review supporting materials
Phase 2: Evaluation (30-60 min)
- Score each of 12 dimensions (1-10)
- Provide evidence for each score
- Identify strengths and weaknesses
- Compare against industry leaders
Phase 3: Synthesis (20 min)
- Create comprehensive report
- Prioritize improvements
- Provide actionable recommendations
- Benchmark against standards
Total Time: 60-90 minutes for thorough audit
Scoring System
1-10 Scale
- 10 - Exceptional, industry-leading
- 9 - Excellent, significantly exceeds expectations
- 8 - Very good, above average
- 7 - Good, meets expectations
- 6 - Acceptable, minimum standards
- 5 - Below average, needs improvement
- 4 - Poor, major gaps
- 3 - Very poor, fundamental problems
- 2 - Critical issues
- 1 - Non-functional
Overall Score
Weighted average of all dimensions:
Overall = (CodeQuality×10% + Architecture×10% + Documentation×10% +
Usability×10% + Performance×8% + Security×10% + Testing×8% +
Maintainability×8% + DevExp×10% + Accessibility×8% +
CI/CD×5% + Innovation×3%)
Usage Examples
Example 1: Audit AI Dev Standards
You: "Use the quality-auditor skill to evaluate ai-dev-standards"
Claude:
"I'll conduct a comprehensive quality audit of ai-dev-standards
across all 12 dimensions.
Phase 1: Discovery...
[Reads documentation, examines code, tests functionality]
Phase 2: Evaluation...
[Scores each dimension with evidence]
Phase 3: Report...
[Generates comprehensive audit report]
# Quality Audit Report: ai-dev-standards
Overall Score: 8.7/10 - Excellent
[Full detailed report with scores, evidence, recommendations]"
Example 2: Compare Tools
You: "Use the quality-auditor skill to compare Next.js vs Remix"
Claude:
"I'll audit both frameworks across all 12 dimensions and provide
a comparative analysis.
[Evaluates both]
# Comparative Quality Audit
| Dimension | Next.js | Remix | Winner |
|-----------|---------|-------|--------|
| Code Quality | 9/10 | 8/10 | Next.js |
| Documentation | 10/10 | 8/10 | Next.js |
| Performance | 8/10 | 9/10 | Remix |
...
[Full comparison with recommendations]"
Example 3: Pre-Launch Quality Gate
You: "Use the quality-auditor skill to check if my app is production-ready"
Claude:
"I'll conduct a pre-launch audit focusing on critical dimensions
for production deployment.
# Pre-Launch Audit Report
**Production Readiness:** Not Yet ⚠️
**Blocking Issues:**
1. Security: 4/10 - Critical vulnerabilities found
2. Testing: 3/10 - Only 15% coverage
3. Performance: 5/10 - Page load >5s
**Must Fix Before Launch:**
[Prioritized list of critical issues]
**Timeline:** 2-3 weeks to address blocking issues"
Report Structure
Executive Summary
- Overall score + rating
- Top 3 strengths
- Top 3 weaknesses
- Final recommendation
Detailed Scores
- Table with all 12 dimensions
- Individual scores + ratings
- Priority levels
Dimension Analysis
- Each dimension examined
- Strengths + weaknesses
- Evidence cited
- Specific improvements
Recommendations
- Immediate - Quick wins (1-2 weeks)
- Short-term - Important improvements (1-3 months)
- Long-term - Strategic changes (3-12 months)
Comparative Analysis
- Industry leader comparison
- Benchmark against similar tools
- Unique differentiators
Risk Assessment
- High/medium/low risk issues
- Impact analysis
- Mitigation strategies
Industry Standards Referenced
The skill compares against established standards:
Code Quality:
- Clean Code (Robert Martin)
- Code Complete (Steve McConnell)
- SonarQube quality gates
Security:
- OWASP Top 10
- SANS Top 25
- CWE/SANS
Accessibility:
- WCAG 2.1 (AA/AAA)
- ADHD-friendly design principles
Testing:
- Test Pyramid (Mike Cohn)
- 80% minimum coverage
Performance:
- Core Web Vitals (Google)
- RAIL model
Special Features
ADHD-Friendly Evaluation
Additional scoring for ADHD-friendliness:
- One-command simplicity (10/10 = single command)
- Automatic everything (10/10 = zero manual steps)
- Clear visual feedback (progress indicators)
- Minimal decisions (sensible defaults)
- Forgiving design (easy undo, backups)
- Low cognitive load (simple mental model)
Developer Experience Focus
Special attention to DX:
- Setup time (<5 min = excellent)
- Error message quality
- Debugging experience
- Documentation clarity
- Tool support
Security-First
Comprehensive security audit:
- OWASP Top 10 check
- Dependency vulnerabilities
- Authentication/authorization review
- Data encryption assessment
- Secret management audit
Benchmarking
Industry Leaders Compared
Code Quality: Linux kernel, SQLite
Documentation: Stripe API, Tailwind CSS
Usability: Vercel, Netlify
Developer Experience: Next.js, Vite
Performance: Cloudflare Workers, Deno
Security: 1Password, Signal
Key Principles
- Be Rigorous - Compare against the best
- Be Objective - Evidence-based only
- Be Constructive - Actionable improvements
- Be Comprehensive - All 12 dimensions
- Be Honest - No inflated scores
- Be Specific - Cite examples
- Be Actionable - Implementable recommendations
Limitations
What this skill does NOT do:
- ❌ Automated code analysis (use tools like SonarQube)
- ❌ Security penetration testing (use Burp Suite, OWASP ZAP)
- ❌ Performance load testing (use k6, JMeter)
- ❌ Replace human judgment (it augments, not replaces)
This skill provides:
- ✅ Comprehensive manual review
- ✅ Expert evaluation framework
- ✅ Structured assessment
- ✅ Actionable recommendations
Tips for Best Results
1. Provide Context
❌ "Audit this tool"
✅ "Audit ai-dev-standards, focusing on developer experience and ADHD-friendliness"
2. Be Specific About Goals
❌ "Tell me if it's good"
✅ "Is this production-ready? What are the blocking issues?"
3. Request Comparisons
❌ "What's the score?"
✅ "How does this compare to industry leaders like Vercel?"
4. Ask for Priorities
❌ "List all issues"
✅ "What are the top 3 issues I should fix first?"
Sample Report
See EXAMPLES/quality-audit-report-sample.md for a complete example report.
Related Skills
- performance-optimizer - For detailed performance analysis
- security-engineer - For security-specific audits
- testing-strategist - For testing strategy
- technical-writer - For documentation improvement
Frequently Asked Questions
Q: How long does an audit take?
A: 60-90 minutes for a thorough audit of a medium-sized project. Larger projects may take 2-3 hours.
Q: Can I audit my own project?
A: Yes! This is actually recommended. Regular self-audits help maintain quality.
Q: How often should I audit?
A:
- Pre-launch: Always
- Major releases: Every time
- Ongoing: Quarterly or bi-annually
- After incidents: When issues occur
Q: What if I disagree with a score?
A: Scores are evidence-based. If you disagree, provide counter-evidence and request re-evaluation of that dimension.
Q: Can this audit closed-source tools?
A: Partially. Without code access, code quality and architecture scores are limited. Documentation, usability, and performance can still be evaluated.
Q: Does a high score guarantee success?
A: No. Quality is necessary but not sufficient. Market fit, timing, and execution also matter. But high quality significantly increases chances of success.
Changelog
v1.0.0 (2025-10-22)
- Initial release
- 12-dimension evaluation framework
- Comprehensive scoring rubric
- Industry standards benchmarking
- ADHD-friendly special considerations
Contributing
Found a missing evaluation criterion? Suggest improvements:
- Identify the gap
- Provide rationale
- Suggest scoring rubric
- Reference industry standards
License
Part of ai-dev-standards. See main LICENSE file.
Ready to evaluate?
Use the quality-auditor skill to audit [tool-name]
Or get specific:
Use the quality-auditor skill to:
1. Audit ai-dev-standards
2. Focus on developer experience and ADHD-friendliness
3. Compare against industry leaders
4. Provide top 5 improvement priorities
Built to measure excellence in software development 📊