code-review

name	code-review
description	Comprehensive code review combining code quality and performance analysis

/code-review — Full Code Review

You are executing the /code-review command. Run code-reviewer and performance-profiler agents in parallel for comprehensive review.

Parallel Execution

Launch both agents simultaneously:

Agent 1: code-reviewer

Review the specified code for:

1. Re-render issues — Unnecessary renders, missing memoization, unstable references
2. React Native anti-patterns — Direct style mutations, ScrollView for long lists, Animated API usage
3. Platform parity — iOS/Android behavioral differences, platform-specific bugs
4. Expo SDK validation — Correct module usage, deprecated APIs, config issues
5. Accessibility — Missing labels, touch target sizes, screen reader support
6. Security — Hardcoded secrets, insecure storage, unvalidated deep links

Agent 2: performance-profiler

Analyze for performance issues:

1. Rendering — Component render counts, unnecessary re-renders
2. Bundle size — Large imports, tree-shaking opportunities
3. Memory — Listener cleanup, large object retention
4. Animations — JS thread animations, Reanimated opportunities

Output Format

Combine results from both agents, grouped by severity:

## Critical (must fix before merge)
[Issues from both agents]

## Warnings (should fix)
[Issues from both agents]

## Suggestions (nice to have)
[Issues from both agents]

## Positive Observations
[Good patterns found]

Notes

• If agent-device is available, take screenshots to verify UI rendering
• Apply rules from rules/common/ and active platform layer
• Flag any violations of project rules