Review

Perform a comprehensive code review of recent changes.

Workflow

1. Gather Changes

bash

git status
git diff --name-only HEAD~5
git log --oneline -10

Categorize modified files by type (check, command, core, tests, docs).

2. Review Against Project Standards

Python code: Type hints on public functions? Async patterns correct? Pydantic models proper? Error handling appropriate?

Checks (iam_validator/checks/): Inherits PolicyCheck? Has check_id, description, default_severity ClassVars? Uses self.get_severity(config)? Returns list[ValidationIssue]?

Tests: Uses @pytest.mark.asyncio? Mock fixtures for AWS calls? Both positive and negative cases? Config overrides tested?

Security: No secrets/credentials? No overly permissive file operations? Input validation present?

Documentation: CHANGELOG.md updated? CLAUDE.md files updated if structure changed? Commits signed?

3. Run Quality Checks

bash

uv run ruff check iam_validator/
uv run mypy iam_validator/ --ignore-missing-imports
uv run pytest tests/ -v --tb=short

4. Output Format

code

### Summary
- X files reviewed
- Y issues found (X critical, Y high, Z medium)

### Critical Issues
- [file:line] Description and fix

### High Issues
- [file:line] Description and fix

### Recommendations
- General improvements

### Tests
- Pass/fail status
- Coverage changes