Comprehensive Code Review Assistant

Purpose

Orchestrate multi-agent swarm review of pull requests with specialized reviewers for security, performance, style, test coverage, and documentation. Provides detailed feedback with auto-fix suggestions and merge readiness assessment.

Core Principles

• Multi-Agent Specialization: Dedicated agents for each review dimension
• Parallel Analysis: Concurrent review across all quality vectors
• Evidence-Based: Measurable quality metrics and validation gates
• Auto-Fix Capability: Automated corrections where possible
• Merge Readiness: Clear approval/rejection criteria

Phase 1: Security Review

Objective

Identify and report security vulnerabilities, OWASP violations, and authentication/authorization issues.

Agent Configuration

agent: security-manager
specialization: security-audit
validation: OWASP-Top-10

Execution Steps

1. Initialize Security Scan

# Pre-task setup
npx claude-flow@alpha hooks pre-task \
  --agent-id "security-manager" \
  --description "Security vulnerability scanning" \
  --task-type "security-audit"

# Restore session context
npx claude-flow@alpha hooks session-restore \
  --session-id "code-review-swarm-${PR_ID}" \
  --agent-id "security-manager"

2. OWASP Top 10 Scan

# Scan for OWASP vulnerabilities
npx eslint . --format json --config .eslintrc-security.json > security-report.json

# Check for dependency vulnerabilities
npm audit --json > npm-audit.json

# Scan for secrets and credentials
npx gitleaks detect --source . --report-path gitleaks-report.json

3. Authentication/Authorization Review

// Analyze authentication patterns
const authPatterns = {
  jwt_validation: /jwt\.verify\(/g,
  password_hashing: /bcrypt|argon2|scrypt/g,
  sql_injection: /\$\{.*\}/g,
  xss_prevention: /sanitize|escape|DOMPurify/g,
  csrf_protection: /csrf|csurf/g
};

// Validate security controls
const securityChecks = {
  has_jwt_validation: false,
  has_password_hashing: false,
  has_sql_parameterization: false,
  has_xss_prevention: false,
  has_csrf_protection: false
};

4. Store Security Findings

# Store results in memory
npx claude-flow@alpha hooks post-edit \
  --file "security-report.json" \
  --memory-key "swarm/security-manager/findings" \
  --metadata "{\"critical\": ${CRITICAL_COUNT}, \"high\": ${HIGH_COUNT}}"

5. Generate Security Report

## Security Review Results

### Critical Issues (Blocking)
- [ ] SQL injection vulnerability in user.controller.js:45
- [ ] Hardcoded API key in config/production.js:12

### High Priority Issues
- [ ] Missing JWT expiration validation
- [ ] Weak password hashing (MD5 detected)

### Recommendations
1. Implement parameterized queries for all database operations
2. Move credentials to environment variables
3. Add JWT expiration checks (max 1 hour)
4. Upgrade to bcrypt with work factor 12+

6. Post-Task Cleanup

npx claude-flow@alpha hooks post-task \
  --task-id "security-review-${PR_ID}" \
  --status "complete" \
  --metrics "{\"issues_found\": ${TOTAL_ISSUES}, \"critical\": ${CRITICAL_COUNT}}"

Validation Gates

• ✅ No critical security issues
• ✅ All OWASP Top 10 checks pass
• ✅ No hardcoded secrets detected
• ✅ Authentication properly implemented

Expected Outputs

• security-report.json - Detailed security findings
• security-score.json - Security quality metrics (0-100)
• auto-fix-suggestions.json - Automated fix recommendations

Phase 2: Performance Review

Objective

Analyze code efficiency, detect bottlenecks, and recommend performance optimizations.

Agent Configuration

agent: performance-analyzer
specialization: performance-audit
metrics: latency, memory, cpu

Execution Steps

1. Initialize Performance Analysis

npx claude-flow@alpha hooks pre-task \
  --agent-id "performance-analyzer" \
  --description "Performance bottleneck analysis" \
  --task-type "performance-audit"

2. Static Performance Analysis

// Detect anti-patterns
const performanceAntiPatterns = {
  nested_loops: /for.*for.*for/gs,
  blocking_operations: /fs\.readFileSync|execSync/g,
  memory_leaks: /global\.|window\./g,
  inefficient_queries: /SELECT \*/gi,
  n_plus_one: /\.map\(.*await/gs
};

// Analyze complexity
const complexityMetrics = {
  cyclomatic_complexity: [], // McCabe complexity
  cognitive_complexity: [],   // Sonar cognitive
  nesting_depth: [],          // Max nesting level
  function_length: []         // Lines per function
};

3. Runtime Performance Profiling

# Run performance benchmarks
npm run test:perf -- --profile

# Analyze bundle size
npx webpack-bundle-analyzer stats.json --mode static -r bundle-analysis.html

# Check memory usage
node --inspect --expose-gc performance-test.js

4. Bottleneck Detection

// Identify slow operations
const bottlenecks = [
  {
    file: "api/users.controller.js",
    function: "getUsersList",
    issue: "N+1 query pattern",
    current_latency: "1250ms",
    optimized_latency: "45ms",
    fix: "Use JOIN or eager loading"
  },
  {
    file: "utils/processor.js",
    function: "processData",
    issue: "Synchronous file operations",
    current_latency: "800ms",
    optimized_latency: "120ms",
    fix: "Use fs.promises.readFile"
  }
];

5. Generate Performance Report

## Performance Review Results

### Critical Bottlenecks (P0)
- **N+1 Query Pattern** (api/users.controller.js:67)
  - Impact: 1250ms → 45ms (27.7x improvement)
  - Fix: Replace `.map(await)` with single JOIN query

### Performance Metrics
| Metric | Current | Target | Status |
|--------|---------|--------|--------|
| API Response Time | 850ms | <200ms | ⚠️ FAIL |
| Memory Usage | 512MB | <256MB | ⚠️ FAIL |
| Bundle Size | 2.4MB | <1MB | ⚠️ FAIL |
| Lighthouse Score | 62 | >90 | ⚠️ FAIL |

### Optimization Recommendations
1. Implement query result caching (Redis)
2. Use async/await for file operations
3. Code splitting for bundle size reduction
4. Lazy load non-critical components

6. Store Performance Data

npx claude-flow@alpha hooks post-edit \
  --file "performance-report.json" \
  --memory-key "swarm/performance-analyzer/metrics" \
  --metadata "{\"bottlenecks\": ${BOTTLENECK_COUNT}, \"avg_latency\": ${AVG_LATENCY}}"

Validation Gates

• ✅ API response time <200ms
• ✅ Memory usage <256MB
• ✅ No O(n²) or worse algorithms
• ✅ Bundle size <1MB

Expected Outputs

• performance-report.json - Performance analysis
• bottlenecks.json - Identified slow operations
• optimization-plan.json - Recommended fixes

Phase 3: Style Review

Objective

Verify code conventions, linting rules, and consistent formatting across the codebase.

Agent Configuration

agent: code-review-swarm
specialization: style-audit
standards: ESLint, Prettier, TypeScript

Execution Steps

1. Initialize Style Check

npx claude-flow@alpha hooks pre-task \
  --agent-id "code-review-swarm" \
  --description "Code style and conventions audit" \
  --task-type "style-audit"

2. Run Linting Tools

# ESLint for JavaScript/TypeScript
npx eslint . --ext .js,.jsx,.ts,.tsx --format json > eslint-report.json

# Prettier for formatting
npx prettier --check "**/*.{js,jsx,ts,tsx,json,css,md}" --list-different > prettier-report.txt

# TypeScript compiler for type checks
npx tsc --noEmit --pretty false 2> typescript-errors.txt

3. Analyze Naming Conventions

// Check naming patterns
const namingConventions = {
  classes: /^[A-Z][a-zA-Z0-9]*$/,        // PascalCase
  functions: /^[a-z][a-zA-Z0-9]*$/,      // camelCase
  constants: /^[A-Z][A-Z0-9_]*$/,        // UPPER_SNAKE_CASE
  components: /^[A-Z][a-zA-Z0-9]*$/,     // PascalCase (React)
  files: /^[a-z][a-z0-9-]*\.(js|ts)$/   // kebab-case
};

// Validate naming compliance
const namingVio