← Back to Skills Marketplace
axelhu

Superpowers Receiving Code Review

by AxelHu · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
193
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install openclaw-receiving-code-review
Description
Use when receiving code review feedback - requires technical verification before implementing suggestions, with reasoned pushback when feedback is technicall...
README (SKILL.md)

Superpowers Receiving Code Review

概述

代码审查需要技术评估,不是情感表演。

核心原则: 实现前验证。有疑问先问。技术正确性 > 社交舒适度。

响应模式

收到代码审查反馈时:

1. 读:完整反馈,不要立刻反应
2. 理解:用自己话复述要求(或提问)
3. 验证:对照代码库现实检查
4. 评估:对本代码库在技术上是合理的吗?
5. 响应:技术确认或合理反驳
6. 实现:一次一个,测试每个

禁止的回应

永远不要:

  • "你说得对!"(明确违反准则)
  • "好观点!" / "优秀的反馈!"(表演性的)
  • "我现在来实现"(验证前不要行动)

改为:

  • 复述技术要求
  • 提问澄清
  • 有技术理由地反驳
  • 直接开始工作(行动 > 言语)

处理不清晰的反馈

如果任何项不清晰:
  停止——暂不实现任何内容
  提问澄清

原因:项可能相关。 partial 理解 = 错误实现。

来源特定处理

来自主人

  • 可信 — 理解后实现
  • 范围不清也问
  • 不要表演性同意
  • 直接行动或技术确认

来自外部审查者

实现前检查:
  1. 技术上对本代码库正确吗?
  2. 破坏现有功能吗?
  3. 当前实现的原因是什么?
  4. 在所有平台/版本上工作吗?
  5. 审查者理解完整上下文吗?

如果建议看起来不对:
  用技术理由反驳

如果无法轻易验证:
  说出来:"没有 X 我无法验证。应该 [调查/提问/继续]?"

YAGNI 检查"专业"功能

如果审查者建议"正确实现"某功能:
  grep 代码库找实际用法

  如果未使用:"此端点没有被调用。YAGNI 移除?"
  如果使用了:那就正确实现

实现顺序

多项目反馈:
  1. 先澄清任何不清晰的项
  2. 然后按此顺序实现:
     - 阻塞问题(破坏、安全)
     - 简单修复(typo、import)
     - 复杂修复(重构、逻辑)
  3. 每个单独测试
  4. 验证无回归

何时反驳

在以下情况反驳:

  • 建议破坏现有功能
  • 审查者缺乏完整上下文
  • 违反 YAGNI(未使用的功能)
  • 对本技术栈在技术上不正确
  • 存在遗留/兼容性原因
  • 与主人的架构决策冲突

如何反驳:

  • 用技术理由,不是防御
  • 问具体问题
  • 引用工作的测试/代码
  • 如涉及架构则让主人参与

承认正确反馈

当反馈正确时:

✅ "已修复。[简要描述变更。]"
✅ "好发现——[具体问题]。已在 [位置] 修复。"

❌ "你说得太对了!"
❌ "好观点!"
❌ "谢谢纠正!"

为什么不说谢谢: 行动说话。直接修。代码本身表明听到了反馈。

常见错误

错误 修复
表演性同意 陈述要求或直接行动
未验证就实现 先对照代码库检查
不测试就批量实现 一次一个,每个测试
假设审查者对 检查是否破坏东西
回避反驳 技术正确性 > 舒适度
partial 实现 澄清所有项后再实现
无法验证仍继续 说明限制,询问方向
Usage Guidance
This skill is coherent and appears safe: it mainly prescribes how to read, verify, and respond to code-review feedback and expects access to the codebase. Before enabling it, confirm the agent has only the intended repository/workspace access (read/write as appropriate) and that your team's communication norms match the skill's strict discouragement of polite acknowledgements—some teams expect explicit thanks or different phrasing. If you want stricter controls, restrict the skill's invocation to specific repositories or require manual invocation rather than autonomous invocation. If you need a different tone/behavior, request a variant that preserves politeness or makes the phrasing configurable.
Capability Analysis
Type: OpenClaw Skill Name: openclaw-receiving-code-review Version: 1.0.0 The skill bundle contains behavioral instructions (SKILL.md) for an AI agent on how to handle code review feedback professionally. It emphasizes technical verification, testing, and avoiding 'performative' agreement (e.g., 'Good point!') in favor of technical correctness. There is no executable code, no network activity, and no evidence of malicious intent or data exfiltration.
Capability Assessment
Purpose & Capability
Name and description match the SKILL.md: guidance for handling code review feedback. The skill requests no binaries, environment variables, or installs that would be out of scope for this purpose.
Instruction Scope
The runtime instructions explicitly direct the agent to read and verify feedback against the codebase (e.g., 'grep the codebase', '对照代码库现实检查', '引用工作的测试/代码'). That is appropriate for a code-review helper, but it implies access to repository files and history. The skill does not instruct contacting external endpoints or accessing secrets, and its scope is prescriptive (what to say/do).
Install Mechanism
There is no install spec and no code files. This instruction-only approach minimizes installation risk.
Credentials
No environment variables, credentials, or config paths are requested. The SKILL.md does not reference secrets or unrelated services.
Persistence & Privilege
always is false and the skill does not request persistent system presence or modify other skills. The default ability for the model to invoke the skill autonomously is unchanged; that is normal and not problematic here.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install openclaw-receiving-code-review
  3. After installation, invoke the skill by name or use /openclaw-receiving-code-review
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release: OpenClaw-adapted skill.
Metadata
Slug openclaw-receiving-code-review
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Superpowers Receiving Code Review?

Use when receiving code review feedback - requires technical verification before implementing suggestions, with reasoned pushback when feedback is technicall... It is an AI Agent Skill for Claude Code / OpenClaw, with 193 downloads so far.

How do I install Superpowers Receiving Code Review?

Run "/install openclaw-receiving-code-review" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Superpowers Receiving Code Review free?

Yes, Superpowers Receiving Code Review is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Superpowers Receiving Code Review support?

Superpowers Receiving Code Review is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Superpowers Receiving Code Review?

It is built and maintained by AxelHu (@axelhu); the current version is v1.0.0.

💬 Comments