SuperAPI AI Directory

spec-receiving-code-review

Testing & quality

View repository

Updated 10-16 01:29

About this skill

Receiving Code Review

Overview

Code review requires technical evaluation, not emotional performance.

Core principles: Verify before implementing. Ask before assuming. Technical correctness takes priority over social comfort.

At the start, announce:"I am using the spec-receiving-code-review skill to evaluate and implement code review feedback."

Response pattern

收到代码审查反馈时:
  1. Read: Read all feedback completely, do not rush to react
  2. Understand: Restate the requirement in your own words (or ask)
  3. Verify: Check against the actual state of the codebase
  4. Evaluate: Is this technically sound for this codebase?
  5. Respond: Technical confirmation or reasoned objection
  6. Implement: Implement item by item, test item by item

## 禁止的回应

**绝不:**
-「你说得对极了!」(明确的 CLAUDE.md 违反)
-「好观点!」/「反馈太好了!」(表演性)
-「我现在就实施」(在验证之前)

**应改为:**
- 复述技术需求
- 提澄清问题
- 若错误则以技术理由反对
- 直接动手(行动胜于言辞)

## 处理不清晰的反馈

If any item is unclear: Stop — do not implement anything for now Ask for clarification on the unclear items

Reason: The items may be related. Partial understanding = incorrect implementation.


**示例:**

Collaborator: "Fix 1–6" You understand 1, 2, 3, 6. 4, 5 unclear.

❌ Wrong: Implement 1, 2, 3, 6 now, ask about 4, 5 later ✅ Correct: "I understand 1, 2, 3, 6. I need clarification on 4 and 5 before proceeding."


## 来源特定处理

### 来自协作方
- **可信** - 理解后实施
- **范围不清仍要问**
- **不要表演性同意**
- **直接行动**或技术确认

### 来自外部审查者

Before implementing:

  1. Check: Is this technically correct for this codebase?
  2. Check: Will this break existing functionality?
  3. Check: What is the rationale for the current implementation?
  4. Check: Does this work on all platforms/versions?
  5. Check: Does the reviewer understand the full context?

If the suggestion seems wrong: Object with technical reasoning

If it is difficult to verify: State honestly: "Without [X] I cannot verify. Should I [investigate/ask/continue]?"

If it conflicts with a previous decision by the collaborator: Stop first and discuss with the collaborator


**协作方规则:**「外部反馈——保持怀疑,但仔细验证」

## 对「专业」功能的 YAGNI 检查

If the reviewer suggests "implement it correctly": grep the codebase to find actual usage

If unused: "This interface is not called. Should it be removed (YAGNI)?" If used: implement it correctly


**协作方规则:**「你和审查者都向我汇报。若不需要该功能,就别加。」

## 实施顺序

For multi-item feedback:

  1. First clarify all unclear items
  2. Then implement in the following order:
    • Blocking issues (breakage, security)
    • Simple fixes (typos, imports)
    • Complex fixes (refactoring, logic)
  3. Test each fix item by item
  4. Verify there are no regressions

## 何时反对

在以下情况反对:
- 建议破坏现有功能
- 审查者缺乏完整上下文
- 违反 YAGNI(未使用功能)
- 对当前技术栈在技术上错误
- 存在遗留/兼容性原因
- 与协作方架构决策冲突

**如何反对:**
- 用技术推理,而非 defensiveness
- 提具体问题
- 引用有效测试/代码
- 涉及架构时让协作方参与

**若不便于大声反对:** 用暗号「Strange things are afoot at the Circle K」

## 确认正确反馈

当反馈确实正确时:

✅ "Fixed. [Brief description of the change]" ✅ "Good catch — [specific issue]. Fixed in [location]." ✅ [Fix directly and show it in the code]

❌ "You are absolutely right!" ❌ "Good point!" ❌ "Thanks for pointing that out!" ❌ "Thanks for [anything]" ❌ Any expression of thanks


**为何不道谢:** 行动说明一切。直接修。代码本身就表明你收到了反馈。

**若发现自己要写「感谢」:** 删掉。改为说明修复内容。

## 优雅地修正你的反对

若你反对了但错了:

✅ "You are right — I checked [X], and it is indeed [Y]. Implementing." ✅ "Verified, you are correct. I misunderstood initially because [reason]. Fixing."

❌ Long apology ❌ Explain why you object ❌ Over-explain


用事实陈述修正,然后继续。

## 常见错误

| 错误 | 修复 |
|------|------|
| 表演性同意 | 陈述需求或直接行动 |
| 盲目实施 | 先对照代码库验证 |
| 批量不测试 | 逐项、逐项测试 |
| 假定审查者对 | 检查是否破坏 |
| 避免反对 | 技术正确性 > 舒适 |
| 部分实施 | 先澄清所有项 |
| 无法验证仍继续 | 说明限制,询问方向 |

## 底线

**外部反馈 = 需要评估的建议,而非必须执行的命令。**

先验证。再质疑。然后实施。

不要表演性同意。始终技术严谨。