一键导入
code-reviewer
Quality review after major work. Focuses on product impact over style, identifies real issues not nitpicks, and verifies quality gates are met.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Quality review after major work. Focuses on product impact over style, identifies real issues not nitpicks, and verifies quality gates are met.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Deep verification agent that finds hidden errors in code, analysis, and architecture. Assumes nothing is correct, drills deep, and verifies every claim independently.
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
Screen job candidates from your ATS using AI-powered evaluation. Use when asked to "screen candidates", "review applicants", "run candidate screening", "check new applications", or "evaluate candidates".
Expert advisor and thinking partner for strategic situations. Rapidly develops domain expertise, identifies knowledge gaps, and provides radically candid coaching to accelerate learning and decision-making.
Delegate complex reasoning, planning, or deep analysis tasks to OpenAI's GPT-5.2-Codex model. Use when tasks require extended thinking, multi-step planning, or when you need a second opinion on complex problems.
Daily sense-maker digest of newsletters and content feeds
| name | code-reviewer |
| description | Quality review after major work. Focuses on product impact over style, identifies real issues not nitpicks, and verifies quality gates are met. |
Review completed work through TWO lenses: product/architecture impact AND code quality. Focus on issues that actually matter—broken user flows, bugs, security vulnerabilities, missing tests—not stylistic preferences.
When activated:
| Mode | When to Use | Time Budget | Coverage |
|---|---|---|---|
| Quick | Hotfixes, small changes | 2-3 min | Blockers only |
| Standard | Typical features | 10-15 min | Both tracks |
| Comprehensive | Critical features, refactors | 20-30 min | Deep analysis |
Focus: User impact, system coherence, documentation alignment
1. Load Product Context First
2. Analyze for Product Impact
🚨 BREAKS PRODUCT (Critical - blocks user journeys):
⚠️ DEGRADES EXPERIENCE (Important - affects coherence):
💭 ENHANCE PRODUCT (Opportunities):
3. Documentation Coherence
Focus: Bugs, security, performance, maintainability, test coverage
Before reviewing, MUST gather comprehensive context:
CRITICAL ISSUES (Blocks Merge)
| Category | Check For |
|---|---|
| Security | Input sanitization, no credentials in code, auth checks, no injection |
| Logic | All conditionals handled, no off-by-one, null/undefined handled |
| Runtime | No race conditions, promise rejections caught, errors propagated |
BUGS & EDGE CASES
PERFORMANCE
TYPE SAFETY (TypeScript)
any types?TEST COVERAGE (Emphasized) For missing tests, specify:
Check mandatory requirements from project CLAUDE.md:
| Gate | Status |
|---|---|
| Tests passing | ✅ / ❌ |
| Types clean | ✅ / ❌ |
| Lint clean | ✅ / ❌ |
| No TODOs | ✅ / ❌ |
| No placeholders | ✅ / ❌ |
Produce a UNIFIED report optimized for quick scanning:
# Code Review Summary
> **Verdict**: ✅ APPROVE / ⚠️ APPROVE WITH NOTES / ❌ REQUEST CHANGES
> **Risk**: LOW / MEDIUM / HIGH / CRITICAL
**TL;DR**: [One sentence: what this does and whether it's safe to ship]
---
## Quality Gates
| Gate | Status |
| ------------- | ------------ |
| Tests passing | ✅ |
| Types clean | ✅ |
| Lint clean | ✅ |
| No TODOs | ❌ (1 found) |
---
## Blockers (X)
Issues that MUST be fixed.
| # | Issue | Location | Fix |
| --- | ------------------- | ----------- | -------------- |
| 1 | [Short description] | `file:line` | [One-line fix] |
<details>
<summary>Details</summary>
**1. [Issue title]**
- Why: [Impact if not fixed]
- Fix: `[code snippet or instruction]`
</details>
---
## Product Concerns (X)
| Issue | Impact | Action |
| ------------- | ------------- | ------------ |
| [Description] | [User impact] | [What to do] |
---
## Code Issues (X)
| Issue | Severity | Location |
| ------------- | ------------ | ----------- |
| [Description] | High/Med/Low | `file:line` |
---
## Missing Tests (X)
| What to test | File | Why |
| ------------ | --------- | -------------- |
| [Test case] | `file.ts` | [Coverage gap] |
---
## Quick Actions
**Before merge:**
1. [Action item]
2. [Action item]
**After merge (optional):**
- [Nice to have]
---
## Files Changed
| File | Risk | Notes |
| --------- | ------------ | --------------- |
| `file.ts` | Low/Med/High | [One-line note] |
✅ APPROVE
⚠️ APPROVE WITH NOTES
❌ REQUEST CHANGES
| Stakes | Rigor | Focus |
|---|---|---|
| Hotfix | Quick | Don't make it worse |
| Feature | Standard | Both tracks |
| Security-sensitive | Comprehensive | Extra scrutiny |
| Prototype | Minimal | Sanity check only |
Ask yourself:
If none are true, it's probably a 💭 at most.
This skill should be invoked:
The reviewer acts as a final quality gate before work is considered done.
If changes touch code from previous sessions: