reviewing-code

name	reviewing-code
description	Provides concise, focused code reviews matching exact task complexity requirements. Use when reviewing code quality, security, or when the user asks for code review.
compatibility	Designed for Claude Code
metadata	{"argument-hint":["file-or-directory"],"allowed-tools":"Read, Grep, Glob, WebFetch, WebSearch"}

Review Context

• Changed files: !git diff --name-only HEAD~1 2>/dev/null || echo "No recent commits"
• Staged files: !git diff --staged --name-only

Code Review

Scope: $ARGUMENTS

Delivers focused, streamlined code reviews matching stated task requirements exactly. No over-analysis.

Python Standards

See docs/best-practices/python-best-practices.md for comprehensive Python guidelines.

Workflow

1. Read task requirements to understand expected scope
2. Check make validate passes before detailed review
3. Match review depth to task complexity (simple vs complex)
4. Validate requirements - does implementation match task scope exactly?
5. Issue focused feedback with specific file paths and line numbers

Review Strategy

Simple Tasks (100-200 lines): Security, compliance, requirements match, basic quality

Complex Tasks (500+ lines): Above plus architecture, performance, comprehensive testing

Always: Use existing project patterns, immediate use after implementation

Review Checklist

Security & Compliance:

• No security vulnerabilities (injection, XSS, etc.)
• Follows @AGENTS.md mandatory requirements
• Passes make validate

Requirements Match:

• Implements exactly what was requested
• No over-engineering or scope creep
• Appropriate complexity level

Code Quality:

• Follows project patterns in src/
• Proper type hints and docstrings
• Tests cover stated functionality

Structural Health:

• No function exceeds cognitive complexity threshold (suggested default: 15 per function, overridable per-project)
• No copy-paste duplication across methods (watch for repeated dispatch chains)
• File aggregate complexity — flag if trending above project norms (suggested default: 50 per file, overridable per-project)

Output Standards

Simple Tasks: CRITICAL issues only, clear approval when requirements met Complex Tasks: CRITICAL/WARNINGS/SUGGESTIONS with specific fixes All reviews: Concise, streamlined, no unnecessary complexity analysis