Code Review Patterns

Code Review Patterns

Overview

Structured patterns for multi-dimensional code review with strict confidence gating. Only issues with >=80% confidence are reported. Zero tolerance for empty catch blocks.

Four Dimensions

Security (weight: 30%)

• Injection vulnerabilities (SQL, XSS, command)
• Authentication and authorization patterns
• Secrets handling (no hardcoded credentials)
• Input validation and sanitization

Quality (weight: 25%)

• Naming conventions and consistency
• Code structure (SRP, cohesion, coupling)
• Error handling completeness
• Type safety (no any escapes)

Performance (weight: 20%)

• Algorithmic complexity (flag O(n^2) or worse)
• Resource and memory leaks
• Database query efficiency (N+1)
• Caching opportunities

Maintainability (weight: 25%)

• Documentation (JSDoc/TSDoc for public APIs)
• Test coverage adequacy
• Readability (function length, nesting depth)
• Tech debt markers (TODO, FIXME)

Confidence Gating

• Only report issues with confidence >= 80%
• Empty catch blocks are always critical (100% confidence)
• Classify: critical, high, medium, low
• Include actionable remediation for each issue

Router Contract

Every review must produce: STATUS, BLOCKING, REQUIRES_REMEDIATION, issue counts.

When to Use

• Code review step in BUILD workflow
• Fix review in DEBUG workflow
• Full REVIEW workflow

Agents Used

• code-reviewer (primary consumer)
• silent-failure-hunter (error handling dimension)