claude-review

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

<EXTREMELY-IMPORTANT> This skill is review orchestration, not code editing.

Non-negotiable rules:

Read the real diff before building the reviewer prompt.
Always use isolated review execution.
Tell the reviewer to report only, never modify code.
Carry forward exclusion lists on repeated rounds so fixed findings are not re-reported.
Verify returned findings before acting on them.

</EXTREMELY-IMPORTANT>

<EXTREMELY-IMPORTANT> 本Skill用于审查编排，而非代码编辑。

不可违背的规则：

在构建审查提示词前，先读取真实的差异内容。
始终使用隔离式审查执行方式。
告知审查工具仅需提交报告，绝不修改代码。
在多轮审查中沿用排除列表，避免已修复的问题被重复报告。
在处理返回的审查结果前，先对其进行验证。

</EXTREMELY-IMPORTANT>

Claude Review

Claude审查

Inputs

输入项

$request

: Optional scope hint such as

last commit

uncommitted

auth focus

, or

round 2

```
$request
```
：可选的范围提示，例如
```
last commit
```
（最后一次提交）、
```
uncommitted
```
（未提交变更）、
```
auth focus
```
（重点审查认证模块）或
```
round 2
```
（第二轮审查）

Goal

目标

Create a focused Claude review that:

derives its scope from the real diff
gives the reviewer the right focus areas
runs in isolation from the main working tree
returns actionable findings without changing code

创建一个聚焦的Claude审查流程，实现：

从真实差异内容中确定审查范围
为审查工具指明正确的重点关注领域
在独立于主工作树的环境中运行审查
返回可执行的审查结果但不修改代码

Step 0: Resolve the review scope

步骤0：确定审查范围

Determine whether the user wants:

full branch review
last-commit review
uncommitted review
targeted review of a subsystem or recent task

Read the relevant diff summary and changed-file list before building the prompt.

If there is nothing to review, stop and say so explicitly.

Success criteria: The review scope is explicit and backed by a real diff.

判断用户需要的审查类型：

全分支审查
最后一次提交审查
未提交变更审查
针对子系统或近期任务的定向审查

在构建提示词前，先读取相关的差异摘要和变更文件列表。

如果没有可审查的内容，需明确告知用户并停止操作。

成功标准：审查范围明确，且有真实的差异内容作为依据。

Step 1: Build a focused review brief

步骤1：构建聚焦的审查简报

Use the diff to identify:

changed files
changed directories or subsystems
likely risk areas such as auth, secrets, concurrency, shelling out, migrations, or tests
issues already fixed in previous rounds that should not be re-reported

The prompt should include:

review base and scope
changed-file list
focused risk areas
explicit report-only instruction
desired output format

Success criteria: The reviewer brief is specific to the actual changes, not generic boilerplate.

利用差异内容识别：

变更的文件
变更的目录或子系统
潜在的风险领域，如认证、密钥、并发处理、外部命令调用、数据迁移或测试代码
前几轮审查中已修复、不应再被报告的问题

提示词应包含：

审查基准和范围
变更文件列表
聚焦的风险领域
明确的"仅报告"指令
期望的输出格式

成功标准：审查简报针对实际变更内容定制，而非通用模板。

Step 2: Launch the isolated reviewer

步骤2：启动隔离式审查工具

Use

Agent

with:

```
subagent_type: general-purpose
```
```
isolation: "worktree"
```
a prompt that says:
- read the full diff
- read surrounding file context
- verify each finding before reporting
- do not make code changes
- report findings as a markdown table:
```
| # | Priority | File:Line | Description | Suggested Fix |
```
- priority levels: P1 (crash/security), P2 (incorrect behavior), P3 (minor)
- if nothing significant is found, say so explicitly

If the review is expected to take a while, background execution is acceptable, but keep the scope tight enough that the reviewer stays focused.

Success criteria: The review runs in an isolated worktree and cannot mutate the active working copy.

使用

Agent

，配置：

```
subagent_type: general-purpose
```
（通用型子代理）
```
isolation: "worktree"
```
（工作树隔离）
提示词需包含：
- 读取完整的差异内容
- 读取文件的上下文内容
- 在报告前验证每个审查结果
- 不得修改代码
- 以Markdown表格形式报告结果：
```
| # | Priority | File:Line | Description | Suggested Fix |
```
- 优先级等级：P1（崩溃/安全问题）、P2（行为异常）、P3（ minor问题）
- 如果未发现重大问题，需明确告知

如果审查预计耗时较长，可采用后台执行方式，但需确保审查范围足够聚焦，避免审查工具偏离重点。

成功标准：审查在隔离的工作树中运行，无法修改当前活跃的工作副本。

Step 3: Parse and summarize findings

步骤3：解析并总结审查结果

Report:

scope reviewed
findings grouped by priority
file and line references
explicit clean result when no material findings are returned

Treat reviewer output as candidate findings, not automatic truth. If the user wants fixes, verify each finding locally first.

Success criteria: The user gets a readable review summary without needing to parse raw agent output.

报告内容包括：

已审查的范围
按优先级分组的审查结果
文件和行号引用
当未发现实质性问题时，明确给出无问题结果

将审查工具的输出视为候选结果，而非绝对事实。如果用户需要修复，需先在本地验证每个审查结果。

成功标准：用户无需解析原始Agent输出，即可获得易读的审查摘要。

Step 4: Iterate only when useful

步骤4：仅在必要时进行迭代审查

On a second or later round:

include previously fixed findings in the exclusion list
tighten the review focus to new or modified areas
avoid rerunning a generic full-branch review if only a few files changed

Success criteria: Each new round looks for genuinely new issues instead of rehashing old ones.

在第二轮及后续审查中：

将之前已修复的问题加入排除列表
将审查重点收紧至新增或修改的区域
如果仅少数文件变更，避免重新运行通用的全分支审查

成功标准：每一轮新审查都聚焦于真正的新问题，而非重复讨论已解决的问题。

Comparison with Sibling Skills

与同类Skill对比

Skill	Tool	Best For
`/claude-review`	Claude Code (Agent)	Deep code understanding, context-aware review
`/codex-review`	OpenAI Codex CLI	Independent perspective, repro scripts
`/kiro-review`	Kiro CLI	Alternative AI perspective

For critical code, run all three and cross-reference findings. The overlap between reviewers catches more bugs than any single tool.

Skill	工具	适用场景
`/claude-review`	Claude Code (Agent)	深度代码理解、上下文感知的审查
`/codex-review`	OpenAI Codex CLI	独立视角、复现脚本生成
`/kiro-review`	Kiro CLI	替代AI视角的审查

对于关键代码，建议同时运行这三个工具并交叉验证结果。多个审查工具的重叠发现比单一工具能捕获更多Bug。

Guardrails

防护规则

Do not add
```
context: fork
```
; this skill already uses the agent runtime directly.
Do not let the reviewer edit code.
Do not skip diff reading before prompt construction.
Do not include real secret values in the prompt or summary.
Do not run this skill proactively; it is explicit-user-only.

请勿添加
```
context: fork
```
；本Skill已直接使用Agent运行时。
禁止让审查工具编辑代码。
构建提示词前，请勿跳过读取差异内容的步骤。
请勿在提示词或摘要中包含真实的密钥值。
请勿主动运行本Skill；仅在用户明确请求时执行。

Output Contract

输出约定

Report:

review scope
changed areas emphasized in the prompt
findings by priority with
```
file:line
```
explicit clean result if no significant findings were returned
whether a later review round should exclude prior fixed issues

报告需包含：

审查范围
提示词中重点强调的变更区域
按优先级排列的审查结果，包含
```
file:line
```
信息
若未发现重大问题，明确给出无问题结果
后续审查是否应排除之前已修复的问题