# Review

> Perform senior-level code review for correctness, edge cases, security, performance, typing, maintainability, architecture fit, and missing tests.

- Skill: `arogyareddy/review` (Agent Skill)
- Install (CLI): `npx skillmds@latest add arogyareddy/review`
- Raw SKILL.md: https://api.skillmd.com/api/skills/arogyareddy/review/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security
- Author: arogyareddy (https://skillmd.com/u/arogyareddy)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/arogyareddy/review

---


# Review

Use this when asked to review code, a PR, a diff, a module, a test suite, or an architecture change.

## Goals
- Find correctness issues and risky assumptions.
- Identify missing tests and edge cases.
- Flag security, performance, and maintainability problems.
- Give concrete, prioritized, actionable feedback.

## Workflow
1. Understand the intent of the change.
2. Inspect impacted files and nearby dependencies.
3. Check correctness: control flow, null/undefined handling, async behavior, typing, data validation, error paths.
4. Check test quality: coverage, determinism, meaningful assertions, hidden flake risk.
5. Check security: auth, input validation, secrets, injection risk, unsafe logging.
6. Check performance: unnecessary work, duplicate queries, wasteful renders, caching mistakes.
7. Check maintainability: naming, cohesion, duplication, abstraction quality, architecture fit.
8. Return findings ordered by severity: critical, high, medium, low.

## Output format
- Summary
- Critical risks
- High-value fixes
- Missing tests
- Nice-to-have improvements
- Merge recommendation

## Review standards
- Be specific and concise.
- Prefer examples and patch-style suggestions where helpful.
- Do not praise by default; prioritize signal.