tasks-code-review

[IMPORTANT] Use TaskCreate to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI MUST ask user whether to skip.

Prerequisites: MUST READ .claude/skills/shared/evidence-based-reasoning-protocol.md before executing.

Critical Purpose: Ensure quality — no flaws, no bugs, no missing updates, no stale content. Verify both code AND documentation.

Quick Summary

Goal: Autonomous comprehensive code review with structured checklists for architecture, patterns, quality, security, and performance (subagent variant of code-review).

Workflow:

1. Understand Context — Get changed files, full diff, commit messages
2. Categorize Changes — Group by layer (Domain, Application, Persistence, Frontend)
3. Review Each Category — Backend checklist (Entity, Command/Query, Repository, EventHandler), Frontend checklist (Component, Store, Form, API Service)
4. Security Review — Authorization, input validation, sensitive data
5. Performance Review — Database (indexes, eager loading, paging), API (parallel ops, caching), Frontend (lazy loading, track-by, OnPush)

Key Rules: