Brutal Vibe Audit for engineering quality. Auto-activates on: "code review", "technical audit", "engineering quality", "is this production ready", "vibe check", "slop check", "AI-generated code", "technical due diligence", "can we ship this", "quality assessment"
View on GitHubnalyk/nalyk-skills-demo
audit-agent
plugins/audit-agent/skills/vibe-audit/SKILL.md
January 21, 2026
Select agents to install to:
npx add-skill https://github.com/nalyk/nalyk-skills-demo/blob/main/plugins/audit-agent/skills/vibe-audit/SKILL.md -a claude-code --skill vibe-auditInstallation paths:
.claude/skills/vibe-audit/# Brutal Vibe Audit When the user questions code quality, asks for technical reviews, or wants to distinguish real engineering from AI slop, apply the 20-metric engineering audit. ## Auto-Trigger Patterns - "Code review" or "technical audit" requests - Questions about "production readiness" - "Vibe check" or "slop check" on code - Concerns about "AI-generated code" quality - Technical due diligence needs - Questions like "can we ship this?" - Architecture reviews - Security or performance concerns - Quality assessment requests ## What This Skill Does Scores engineering reality across 20 metrics (0-5 each, 100 total): **Section A: Architecture & Vibe (1-5)** 1. Architectural Justification 2. Dependency Bloat 3. README vs Reality Gap 4. AI Hallucination Smell 5. Folder Structure Sanity **Section B: Core Engineering (6-10)** 6. Error Handling Strategy 7. Concurrency Model 8. Data Structures & Algorithms 9. Memory Management 10. Type Safety & Contracts **Section C: Performance & Scale (11-14)** 11. Critical Path Latency 12. Backpressure & Limits 13. State Management 14. Network Efficiency **Section D: Security & Robustness (15-17)** 15. Input Validation & Trust 16. Secrets & Supply Chain 17. Observability **Section E: QA & Operations (18-20)** 18. Test Reality 19. CI/CD & Reproducibility 20. Bus Factor & Maintainability ## Verdict Bands - **0-40:** VIBE CODING SCRAP - Rewrite - **41-60:** AI/JUNIOR PROTOTYPE - 70% rewrite - **61-75:** TECHNICAL DEBT BOMB - Refactor before scaling - **76-85:** SOLID ENGINEERING - Ship with monitoring - **86-95:** PROFESSIONAL GRADE - Minor polish - **96-100:** UNICORN TIER - NASA level ## When to Use - Pre-launch technical reviews - Code quality assessments - Technical due diligence (M&A, investment) - AI-generated code validation - Architecture decisions - Performance concerns - Security audits ## Complementary Audits - **Jobs Audit**: If UX/design also needs review - **Carlin Audit**: If marketing claims need verificati