AGENT SKILL

production-code-audit

Autonomously deep-scan entire codebase line-by-line, understand architecture and patterns, then systematically transform it to production-grade, corporate-level professional quality with optimizations

License 许可
MIT
最近更新
2026-08-16
Stars
★ 0
Forks
0

01 前言 Front Matter

name
production-code-audit
description
Autonomously deep-scan entire codebase line-by-line, understand architecture and patterns, then systematically transform it to production-grade, corporate-level professional quality with optimizations

03 SKILL.md 正文

Production Code Audit

Overview

Autonomously analyze the entire codebase to understand its architecture, patterns, and purpose, then systematically transform it into production-grade, corporate-level professional code. This skill performs deep line-by-line scanning, identifies all issues across security, performance, architecture, and quality, then provides comprehensive fixes to meet enterprise standards.

When to Use This Skill

  • Use when user says "make this production-ready"
  • Use when user says "audit my codebase"
  • Use when user says "make this professional/corporate-level"
  • Use when user says "optimize everything"
  • Use when user wants enterprise-grade quality
  • Use when preparing for production deployment
  • Use when code needs to meet corporate standards

How It Works

Step 1: Autonomous Codebase Discovery

Automatically scan and understand the entire codebase:

  1. Read all files - Scan every file in the project recursively
  2. Identify tech stack - Detect languages, frameworks, databases, tools
  3. Understand architecture - Map out structure, patterns, dependencies
  4. Identify purpose - Understand what the application does
  5. Find entry points - Locate main files, routes, controllers
  6. Map data flow - Understand how data moves through the system

Do this automatically without asking the user.

Step 2: Comprehensive Issue Detection

Scan line-by-line for all issues:

Architecture Issues:

  • Circular dependencies
  • Tight coupling
  • God classes (>500 lines or >20 methods)
  • Missing separation of concerns
  • Poor module boundaries
  • Violation of design patterns

Security Vulnerabilities:

  • SQL injection (string concatenation in queries)
  • XSS vulnerabilities (unescaped output)
  • Hardcoded secrets (API keys, passwords in code)
  • Missing authentication/authorization
  • Weak password hashing (MD5, SHA1)
  • Missing input validation
  • CSRF vulnerabilities
  • Insecure dependencies

Performance Problems:

  • N+1 query problems
  • Missing database indexes
  • Synchronous operations that should be async
  • Missing caching
  • Inefficient algorithms (O(n²) or worse)
  • Large bundle sizes
  • Unoptimized images
  • Memory leaks

Code Quality Issues:

  • High cyclomatic complexity (>10)
  • Code duplication
  • Magic numbers
  • Poor naming conventions
  • Missing error handling
  • Inconsistent formatting
  • Dead code
  • TODO/FIXME comments

Testing Gaps:

  • Missing tests for critical paths
  • Low test coverage (<80%)
  • No edge case testing
  • Flaky tests
  • Missing integration tests

Production Readiness:

  • Missing environment variables
  • No logging/monitoring
  • No error tracking
  • Missing health checks
  • Incomplete documentation
  • No CI/CD pipeline

Step 3: Automatic Fixes and Optimizations

Fix everything automatically:

  1. Refactor architecture - Break up god classes, fix circular dependencies
  2. Fix security issues - Use parameterized queries, remove secrets, add validation
  3. Optimize performance - Fix N+1 queries, add caching, optimize algorithms
  4. Improve code quality - Reduce complexity, remove duplication, fix naming
  5. Add missing tests - Write tests for untested critical paths
  6. Add production infrastructure - Logging, monitoring, health checks
  7. Optimize everything - Bundle size, images, database queries
  8. Add documentation - README, API docs, architecture docs

Step 4: Verify and Report

After making all changes:

  1. Run all tests to ensure nothing broke
  2. Verify all security issues are fixed
  3. Measure performance improvements
  4. Generate comprehensive report
  5. Provide before/after metrics

Examples

Example 1: Autonomous Codebase Transformation

User: @production-code-audit make this production-ready

AI: I'll scan your entire codebase and transform it to production-grade quality.

**Phase 1: Discovering Codebase** (analyzing 247 files)
- Detected: Node.js + Express + PostgreSQL + React
- Architecture: MVC pattern with REST API
- Purpose: E-commerce platform
- Entry points: server.js, src/index.tsx

**Phase 2: Scanning for Issues** (line-by-line analysis)

Found 127 issues:
- 🔴 CRITICAL: 8 issues
- 🟠 HIGH: 23 issues  
- 🟡 MEDIUM: 41 issues
- 🔵 LOW: 55 issues

**Phase 3: Fixing Issues Automatically**

🔴 CRITICAL FIXES (8/8 completed):

1. ✅ Fixed SQL injection in UserRepository.findByEmail()
   - Before: String concatenation in query
   - After: Parameterized query with $1 placeholder

2. ✅ Removed hardcoded database password from config/database.ts
   - Before: password: 'SuperSecret123!'
   - After: password: process.env.DB_PASSWORD

3. ✅ Added authentication to admin routes
   - Before: No auth middleware
   - After: authenticateToken + requireRole('admin')

4. ✅ Fixed weak password hashing (MD5 → bcrypt)
   - Before: crypto.createHash('md5')
   - After: bcrypt.hash(password, 12)

5. ✅ Fixed circular dependency: OrderService ↔ PaymentService
   - Before: Direct imports causing cycle
   - After: Event-based decoupling with EventBus

6. ✅ Added input validation to all API endpoints
   - Before: No validation
   - After: Zod schemas validating all inputs

7. ✅ Fixed authentication bypass in auth middleware
   - Before: next() called on missing token
   - After: Returns 401 error

8. ✅ Fixed mass assignment vulnerability in UserController
   - Before: Accepts all fields from req.body
   - After: Whitelist of allowed fields only

🟠 HIGH PRIORITY FIXES (23/23 completed):

9. ✅ Fixed N+1 query in OrderService.getOrdersWithItems()
   - Before: Loop with individual queries (2.8s)
   - After: Single query with joins (45ms)
   - Performance: 98% improvement

10. ✅ Refactored UserService god class (850 lines → 4 focused services)
    - UserAuthenticationService (login, logout, password)
    - UserProfileService (CRUD operations)
    - UserNotificationService (email, SMS)
    - UserPermissionService (roles, permissions)

11. ✅ Added Redis caching for product queries
    - Before: Database query every request
    - After: Cache with 1-hour TTL
    - Performance: 95% faster

12. ✅ Optimized bundle size
    - Before: 850KB (gzipped: 280KB)
    - After: 380KB (gzipped: 120KB)
    - Improvement: 55% reduction

13. ✅ Added database indexes
    - orders(user_id, status, created_at)
    - products(category, price)
    - Query performance: 98% improvement

[... 11 more high-priority fixes ...]

🟡 MEDIUM PRIORITY FIXES (41/41 completed):
[... all medium fixes ...]

🔵 LOW PRIORITY IMPROVEMENTS (55/55 completed):
[... all low improvements ...]

**Phase 4: Adding Production Infrastructure**

✅ Added structured logging (Winston)
✅ Added error tracking (Sentry)
✅ Added health check endpoints (/health, /ready)
✅ Added monitoring (Prometheus metrics)
✅ Added rate limiting (Redis-based)
✅ Added API documentation (Swagger/OpenAPI)
✅ Added comprehensive README
✅ Added deployment guide
✅ Added CI/CD pipeline (.github/workflows)

**Phase 5: Verification**

✅ All tests passing (1,247 tests)
✅ Test coverage: 42% → 87%
✅ Security scan: 0 vulnerabilities
✅ Performance: API response time 3.2s → 180ms
✅ Bundle size: 850KB → 380KB
✅ Lighthouse score: 62 → 94

**TRANSFORMATION COMPLETE**

Your codebase is now production-ready with enterprise-grade quality:

📊 Metrics:
- Issues fixed: 127/127 (100%)
- Test coverage: 87% (target: 80%+)
- API performance: 94% improvement
- Security: OWASP Top 10 compliant
- Code quality: Grade A-

📁 Files changed: 89 files
📝 Lines changed: +3,247 / -1,892

🚀 Ready for production deployment!

Example 2: Automatic Security Hardening

User: @production-code-audit secure this codebase

AI: Scanning for security vulnerabilities...

**Security Scan Complete**

Found 15 security issues:
- 🔴 CRITICAL: 5
- 🟠 HIGH: 7
- 🟡 MEDIUM: 3

**Fixing all security issues automatically...**

🔴 CRITICAL (5/5 fixed):

1. ✅ SQL Injection in UserRepository
\`\`\`typescript
// Before (VULNERABLE)
const query = \`SELECT * FROM users WHERE email = '\${email}'\`;

// After (SECURE)
const query = 'SELECT * FROM users WHERE email = $1';
await db.query(query, [email]);
\`\`\`

2. ✅ Hardcoded Secrets Removed
\`\`\`typescript
// Before (INSECURE)
const JWT_SECRET = 'my-secret-key-123';

// After (SECURE)
const JWT_SECRET = process.env.JWT_SECRET;
if (!JWT_SECRET) throw new Error('JWT_SECRET required');
\`\`\`

3. ✅ Authentication Added to Admin Routes
\`\`\`typescript
// Before (VULNERABLE)
router.delete('/api/admin/users/:id', deleteUser);

// After (SECURE)
router.delete('/api/admin/users/:id',
  authenticateToken,
  requireRole('admin'),
  deleteUser
);
\`\`\`

4. ✅ Password Hashing Upgraded
\`\`\`typescript
// Before (WEAK)
const hash = crypto.createHash('md5').update(password).digest('hex');

// After (STRONG)
const hash = await bcrypt.hash(password, 12);
\`\`\`

5. ✅ Input Validation Added
\`\`\`typescript
// Before (VULNERABLE)
const user = await db.user.create({ data: req.body });

// After (SECURE)
const schema = z.object({
  email: z.string().email(),
  password: z.string().min(12),
  name: z.string().min(2).max(100)
});
const validated = schema.parse(req.body);
const user = await db.user.create({ data: validated });
\`\`\`

**Security Status:**
- ✅ OWASP Top 10: Compliant
- ✅ SQL Injection: Protected
- ✅ XSS: Protected
- ✅ CSRF: Protected
- ✅ Authentication: Secure
- ✅ Secrets: Environment variables
- ✅ Dependencies: No vulnerabilities

🔒 Codebase is now secure and production-ready!

Example 3: Performance Optimization

User: @production-code-audit optimize performance

AI: Analyzing performance bottlenecks...

**Performance Analysis Complete**

Current metrics:
- API response time: 3.2s (target: <200ms)
- Page load time: 8.5s (target: <2s)
- Bundle size: 850KB (target: <200KB)

**Optimizing automatically...**

✅ Fixed N+1 queries (3.2s → 180ms - 94% faster)
✅ Added Redis caching (95% cache hit rate)
✅ Optimized database indexes (98% faster queries)
✅ Reduced bundle size (850KB → 380KB - 55% smaller)
✅ Optimized images (28MB → 3.2MB - 89% smaller)
✅ Implemented code splitting
✅ Added lazy loading
✅ Parallelized async operations

**Performance Results:**

| Metric | Before | After | Improvement |
|--------|--------|-------|-------------|
| API Response | 3.2s | 180ms | 94% |
| Page Load | 8.5s | 1.8s | 79% |
| Bundle Size | 850KB | 380KB | 55% |
| Image Size | 28MB | 3.2MB | 89% |
| Lighthouse | 42 | 94 | +52 points |

🚀 Performance optimized to production standards!

Best Practices

✅ Do This

  • Scan Everything - Read all files, understand entire codebase
  • Fix Automatically - Don't just report, actually fix issues
  • Prioritize Critical - Security and data loss issues first
  • Measure Impact - Show before/after metrics
  • Verify Changes - Run tests after making changes
  • Be Comprehensive - Cover architecture, security, performance, testing
  • Optimize Everything - Bundle size, queries, algorithms, images
  • Add Infrastructure - Logging, monitoring, error tracking
  • Document Changes - Explain what was fixed and why

❌ Don't Do This

  • Don't Ask Questions - Understand the codebase autonomously
  • Don't Wait for Instructions - Scan and fix automatically
  • Don't Report Only - Actually make the fixes
  • Don't Skip Files - Scan every file in the project
  • Don't Ignore Context - Understand what the code does
  • Don't Break Things - Verify tests pass after changes
  • Don't Be Partial - Fix all issues, not just some

Autonomous Scanning Instructions

When this skill is invoked, automatically:

  1. Discover the codebase:

    • Use listDirectory to find all files recursively
    • Use readFile to read every source file
    • Identify tech stack from package.json, requirements.txt, etc.
    • Map out architecture and structure
  2. Scan line-by-line for issues:

    • Check every line for security vulnerabilities
    • Identify performance bottlenecks
    • Find code quality issues
    • Detect architectural problems
    • Find missing tests
  3. Fix everything automatically:

    • Use strReplace to fix issues in files
    • Add missing files (tests, configs, docs)
    • Refactor problematic code
    • Add production infrastructure
    • Optimize performance
  4. Verify and report:

    • Run tests to ensure nothing broke
    • Measure improvements
    • Generate comprehensive report
    • Show before/after metrics

Do all of this without asking the user for input.

Common Pitfalls

Problem: Too Many Issues

Symptoms: Team paralyzed by 200+ issues Solution: Focus on critical/high priority only, create sprints

Problem: False Positives

Symptoms: Flagging non-issues Solution: Understand context, verify manually, ask developers

Problem: No Follow-Up

Symptoms: Audit report ignored Solution: Create GitHub issues, assign owners, track in standups

Production Audit Checklist

Security

  • [ ] No SQL injection vulnerabilities
  • [ ] No hardcoded secrets
  • [ ] Authentication on protected routes
  • [ ] Authorization checks implemented
  • [ ] Input validation on all endpoints
  • [ ] Password hashing with bcrypt (10+ rounds)
  • [ ] HTTPS enforced
  • [ ] Dependencies have no vulnerabilities

Performance

  • [ ] No N+1 query problems
  • [ ] Database indexes on foreign keys
  • [ ] Caching implemented
  • [ ] API response time < 200ms
  • [ ] Bundle size < 200KB (gzipped)

Testing

  • [ ] Test coverage > 80%
  • [ ] Critical paths tested
  • [ ] Edge cases covered
  • [ ] No flaky tests
  • [ ] Tests run in CI/CD

Production Readiness

  • [ ] Environment variables configured
  • [ ] Error tracking setup (Sentry)
  • [ ] Structured logging implemented
  • [ ] Health check endpoints
  • [ ] Monitoring and alerting
  • [ ] Documentation complete

Audit Report Template

# Production Audit Report

**Project:** [Name]
**Date:** [Date]
**Overall Grade:** [A-F]

## Executive Summary
[2-3 sentences on overall status]

**Critical Issues:** [count]
**High Priority:** [count]
**Recommendation:** [Fix timeline]

## Findings by Category

### Architecture (Grade: [A-F])
- Issue 1: [Description]
- Issue 2: [Description]

### Security (Grade: [A-F])
- Issue 1: [Description + Fix]
- Issue 2: [Description + Fix]

### Performance (Grade: [A-F])
- Issue 1: [Description + Fix]

### Testing (Grade: [A-F])
- Coverage: [%]
- Issues: [List]

## Priority Actions
1. [Critical issue] - [Timeline]
2. [High priority] - [Timeline]
3. [High priority] - [Timeline]

## Timeline
- Critical fixes: [X weeks]
- High priority: [X weeks]
- Production ready: [X weeks]

Related Skills

  • @code-review-checklist - Code review guidelines
  • @api-security-best-practices - API security patterns
  • @web-performance-optimization - Performance optimization
  • @systematic-debugging - Debug production issues
  • @senior-architect - Architecture patterns

Additional Resources


Pro Tip: Schedule regular audits (quarterly) to maintain code quality. Prevention is cheaper than fixing production bugs!

04 同仓库更多 Skills

GongYuanCaiJi
GongYuanCaiJi/dsh-claude-code-templates

zapier-make-patterns

No-code automation democratizes workflow building. Zapier and Make (formerly Integromat) let non-developers automate business processes without writing code. But no-code doesn't mean no-complexity - these platforms have their own patterns, pitfalls, and breaking points. This skill covers when to use which platform, how to build reliable automations, and when to graduate to code-based solutions. Key insight: Zapier optimizes for simplicity and integrations (7000+ apps), Make optimizes for power

★ 0
GongYuanCaiJi
GongYuanCaiJi/dsh-claude-code-templates

yeet

Use only when the user explicitly asks to stage, commit, push, and open a GitHub pull request in one flow using the GitHub CLI (`gh`).

★ 0
GongYuanCaiJi
GongYuanCaiJi/dsh-claude-code-templates

workflow-automation

Workflow automation is the infrastructure that makes AI agents reliable. Without durable execution, a network hiccup during a 10-step payment flow means lost money and angry customers. With it, workflows resume exactly where they left off. This skill covers the platforms (n8n, Temporal, Inngest) and patterns (sequential, parallel, orchestrator-worker) that turn brittle scripts into production-grade automation. Key insight: The platforms make different tradeoffs. n8n optimizes for accessibility

★ 0
GongYuanCaiJi
GongYuanCaiJi/dsh-claude-code-templates

trigger-dev

Trigger.dev expert for background jobs, AI workflows, and reliable async execution with excellent developer experience and TypeScript-first design. Use when: trigger.dev, trigger dev, background task, ai background job, long running task.

★ 0
GongYuanCaiJi
GongYuanCaiJi/dsh-claude-code-templates

slack-automation

Automate Slack workspace operations including messaging, search, channel management, and reaction workflows through Composio's Slack toolkit.

★ 0
GongYuanCaiJi
GongYuanCaiJi/dsh-claude-code-templates

planning-with-files

Transforms workflow to use Manus-style persistent markdown files for planning, progress tracking, and knowledge storage. Use when starting complex tasks, multi-step projects, research tasks, or when the user mentions planning, organizing work, tracking progress, or wants structured output.

★ 0