cr
自动代码审查（CR）——只读 subagent 对照规格审查变更，产出问题清单；作为独立审查环节或 /pipe 流水线的一部分
Install / Use
npx skills add zsxink/MusicTagInstalls into whichever agent you are using.
Claude Commands
Claude Code slash commands
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of cr
cr scores 63/100 on our quality scale, 3994th of 4,606 Development & Engineering skills we index.
Its Claude Commands is 1.3 KB long, well organised into 11 sections with 1 code example: moderately detailed.
It has no GitHub stars yet, so there is no community track record; judge it on its content.
Maintenance, license and trust
- The repository was last updated yesterday, so cr is actively maintained.
- No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
- Its trust signals score 80/100, with 2 cautions from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands.
Automated pattern scan on 2026-10-26. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
cr compared with similar skills
All 4 of these similar skills score higher than cr; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| cr (this skill)by zsxink | 63 | 0 | 1d ago | Claude Commands |
| claude-memby thedotmack | 100 | 103.4k | today | CLAUDE.md |
| Agent-Reachby Panniantong | 100 | 103.3k | 1d ago | CLAUDE.md |
| taste-skillby Leonxlnx | 100 | 102.9k | today | CLAUDE.md |
| Understand-Anythingby Egonex-AI | 100 | 88.4k | today | CLAUDE.md |
Frequently asked questions
- How do I install cr?
- Run
npx skills add zsxink/MusicTag. The install tabs above show the steps for each supported agent. - Which AI agents does cr work with?
- It is written for Claude Code, as a Claude Commands file. Other agents that read the same format can often use it too.
- Is cr safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. It declares no license and scores 80/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is cr still maintained?
- The repository was last updated yesterday, so cr is actively maintained.
Skill content
View source on GitHubdescription: 自动代码审查（CR）——只读 subagent 对照规格审查变更，产出问题清单；作为独立审查环节或 /pipe 流水线的一部分
自动代码审查（CR，只读）
对当前分支/变更做一次 subagent 驱动的代码审查。核心原则：CR 只读，不改代码——发现的问题打回 Leader 重新派给开发角色修复。
输入
- 可选：变更名
<name>。省略时从上下文推断或自动选择（唯一 active change 则自动选中）。
审查基准
- 规格权威：
openspec/changes/<name>/（proposal / specs / design / tasks） - 定稿规格：
docs/V1-PRD.md、docs/design/design.md - 审查对象：
git diff main...HEAD（当前分支相对 main 的全部改动）
审查流程
- 定位变更：读
openspec/changes/<name>/下的 specs 与 design，梳理验收点（每个 scenario 是一个潜在测试用例）。 - 生成 diff：
git diff main...HEAD --stat看范围，再读关键文件 diff。 - 派只读审查代理（并行）：
cr-agent代理：对照 specs 审查 规格一致性（实现是否符合）、遗漏（spec 有代码没有）、缺陷（逻辑/边界/测试缺失、与定稿约束冲突）。- 审查代理只读：只用 Bash(读)/Read/Glob/Grep，不 Edit/Write 任何代码。
- 汇总问题清单：按严重度排序（阻断/主要/次要），每条标注：文件、位置、问题、违反的 spec/requirement、建议。
输出
## CR 结果：<change-name>
### 阻断（必须修复）
- [ ] <文件:行> <问题> — 违反 <spec requirement>
### 主要
- [ ] ...
### 次要
- [ ] ...
结论：通过 / 需修复（第 N 轮）
后续（打回机制）
- CR 有阻断/主要问题 → 不打回给 CR 自己修，CR 只读；由 Leader 打回给对应开发角色（Rust-Dev / Vue-Dev）修复 → 重跑受影响测试 → 再复审。
- 三轮未通过 → 挂起：不再自动重试，上报用户决策（改规格 / 人工修复 / 放行）。
- 通过：可进入验证/合并。
Guardrails
- CR 只读：审查代理不 Edit/Write 任何代码，问题经 Leader 中转打回。
- 审查必须对照规格，不能只查代码自身 bug。
- 阻断/主要问题不修复不进入下一步；三轮未通过即挂起，不无限重试。
- 报告如实：无问题就明确说「无阻断、无主要问题」。
Related Skills
claude-mem
103.4kPersistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Agent-Reach
103.3kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
taste-skill
102.9kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
Understand-Anything
88.4kGraphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.