Skip to content
Agentmart
Submit a tool

receive-review

接收并处理审查反馈。逐项核实审查发现,按严重度分类处理,拒绝表演性同意。Use when another agent's review feedback arrives after a code review.

Skill by JianHun101

IndexedMIT
Write a review

Chalk Labs

Need help deploying receive-review?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/jianhun101-receive-review

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

接收并处理审查反馈。逐项核实审查发现,按严重度分类处理,拒绝表演性同意。Use when another agent's review feedback arrives after a code review. Not for review conclusions that are mergeable (nothing to fix), or when you are the reviewer. Output a fix confirmation table with per-item status and test results.

Use cases: testing, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/receive-review

What's inside

Skill name
receive-review
Description
接收并处理审查反馈。逐项核实审查发现,按严重度分类处理,拒绝表演性同意。Use when another agent's review feedback arrives after a code review. Not for review conclusions that are mergeable (nothing to fix), or when you are the reviewer. Output a fix confirmation table with per-item status and test results.
License
MIT
Instructions
121 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# receive-review

接收审查反馈并按严重度分类处理。禁止表演性同意——用代码修改证明你收到了反馈。

## 何时使用

- 收到其他猫的审查反馈时
- 审查(code-review)后审查者返回了结论

## 不使用的情况

- 审查结论是 ✅ 可以合并 / 💬 仅评论(非阻断、无需处理)
- 自己是审查者(用 code-review 返回结果,不是 receive)

## 执行步骤

### 步骤 1:分类审查发现

将审查者的每一条发现归入严重度等级(参照 `skills/refs/review-standards.md`):

- **P1 阻断**:安全漏洞、数据丢失、功能崩溃
- **P2 必须修**:逻辑错误、性能问题、资源泄漏、架构违规
- **P3 可选修**:命名、注释、代码风格

### 双轴处理(code-review 方法论)

审查发现来自两条独立轴,分开处理、分开验证:

- **Standards 发现**(违反文档化规范)→ 对照 CODING_STANDARDS 核实;文档规范明说是硬违规,坏味道是判断标签
- **Spec 发现**(需求缺失/范围蔓延/实现错误)→ 对照发起它的 spec/issue 逐条核实

**一条轴通过不豁免另一条轴**——Standards 全过但 Spec 缺需求仍需修。区分硬违规与判断标签:文档化规范优先于坏味道基线,工具已强制的跳过。

### 步骤 2:VERIFY 三道门(每条发现都要过)

在接受任何审查发现之前:

**门 1 — Spec Gate**:这个修改建议是否与现有需求冲突?
**门 2 — Mechanism Gate**:审查者是否有具体的失败证据,还是仅仅"不够优雅"?
**门 3 — Feature Gate**:采纳这个建议后,核心用户路径是否仍然正常?

三道门通过 → 接受并修复
任何一道门不通过 → 标记为"讨论",说明理由后回复审查者

### 步骤 3:FAILURE-MODE SWEEP(失败模式扫描)

如果 2+ 个 P1/P2 发现属于同一失败模式(同一类 bug 的变体),**在修复任何一条之前**:

1. 扫描整个 diff,查找同一模式的全部出现位置
2. 一次性修复所有同类问题
3. 在修复确认中标注 "sweep: {模式名}, 扫描 N 处, 修复 N 处"

### 步骤 4:逐项修复

按 P1 → P2 → P3 顺序修复:

1. 先修所有 P1
2. 再修所有 P2
3. P3:现场修或明确放弃(写进结论)
4. 每修完一类,运行 `pnpm test` 确认无回归

### 步骤 5:3 轮升级规则

如果同一个有状态对象(同一个文件/模块/函数)连续 3 轮审查都有发现:

- 停止代码级别的修补
- 回到设计层面,重新评估方案
- 可能需要对问题模块做整体重构

### 步骤 6:生成修复确认

```
## 修复确认

| # | 级别 | 发现 | 状态 | 修复 |
|---|------|------|------|------|
| P1-1 | P1 | {描述} | ✅ done | {文件}: {修改说明} |
| P2-1 | P2 | {描述} | ✅ done | {文件}: {修改说明} |
| P3-1 | P3 | {描述} | ❌ dropped | 理由: {为什么不修} |

测试: pnpm test → N passed, 0 failed

@审查者 修复完成,请确认
```

## 禁止的行为

| 禁止                      | 正确做法             |
| ------------------------- | -------------------- |
| "你说得对!" / "好主意!" | 直接用代码修改证明   |
| 盲目接受所有建议          | 每条过 VERIFY 三道门 |
| P3 写 BACKLOG             | P3 现场处理或放弃    |
| 只修一条忽略同类          | Failure-mode sweep   |
| 3 轮后继续修补同一文件    | 升级到设计层面       |

## 衔接

- 修复完成 → 通知审查者确认
- 审查者确认 ✅ → 可以合并
- 审查者仍不满意 → 回到步骤 1(新一轮)
- 3 轮未收敛 → 升级到设计重新评估

## Common Mistakes

| 错误                                 | 正确做法                                                                     |
| ------------------------------------ | ---------------------------------------------------------------------------- |
| 回「你说得对!好主意!」             | 直接用代码修改证明收到了反馈——表演性同意是禁区                               |
| 盲信审查结论不核实                   | 每条发现过 VERIFY 三道门(Spec / Mechanism / Feature),不通过就标记「讨论」 |
| 只修审查者点名的第一条,同类问题漏掉 | 2+ 条同类发现 → failure-mode sweep,扫描全 diff 一次修完                     |
| P3 写进 BACKLOG 攒着                 | P3 现场修或明确放弃,理由写进结论                                            |
| Standards/Spec 两轴混在一起处理      | 分开核实:一条轴过了不代表另一条也过,硬违规与判断标签分开报                 |

## 与其他 skill 区别

| skill        | 区别                                                                         |
| ------------ | ---------------------------------------------------------------------------- |
| code review  | code-review 产生审查结论,receive-review 接收并处理反馈                      |
| quality-gate | quality-gate 是作者自查(提交前);receive-review 是收到他人审查后的处理流程 |
| code-review  | code-review 是执行双轴审查产出结论;receive-review 是结论的接收方            |

From JianHun101/catStudio, MIT.

More from JianHun101/catStudio

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 21, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from JianHun101

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get receive-review running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying receive-review?

CallGet help
Cite this page

receive-review skill. Agent Mart. https://agentmart.dev/skills/jianhun101-receive-review (last updated 2026-10-24)