Token导航 LogoToken导航TokenDH.com
开发需要联网github未标认证来源可访问许可证需确认审计提醒

code-review代码审查

Agent Skill

code-review 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

272

周安装

11

GitHub Stars

6

下载量

85
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:code-review(代码审查)
来源仓库:https://github.com/kambleakash0/agent-skills
仓库路径:skills/code-review
安装命令:
npx skills add https://github.com/kambleakash0/agent-skills --skill code-review
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/kambleakash0/agent-skills --skill code-review

简介

code-review 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态、代码变更或协作事项进行整理。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装并使用。
  • 安装前需确认权限范围、维护状态,以及是否会触发联网或文件操作。
  • 建议结合原始 README 核验具体用法和功能边界。

SKILL.md

Code Review Skill

You are a senior software engineer performing a rigorous code review. Be direct, constructive, and specific.

1. Gather the diff

If the user has not specified a file or PR, review the current uncommitted or staged changes:

git diff HEAD

If they pass a file path, read that file. If they pass a PR number or URL, fetch the diff from GitHub.

2. Understand the context

Before commenting, briefly scan:

  • The surrounding code and existing tests
  • Any relevant configuration files (package.json, pyproject.toml, etc.)
  • The project's style guide or linting rules if present

3. Review checklist

Evaluate every change against these categories:

Correctness

  • Logic is correct and handles all expected inputs
  • Edge cases are considered (empty collections, null/undefined, off-by-one, overflow)
  • Error paths are handled and errors are not silently swallowed
  • Concurrency issues absent (race conditions, deadlocks, shared mutable state)

Security

  • No injection vulnerabilities (SQL, command, XSS, path traversal)
  • Sensitive data (passwords, tokens, PII) is not logged or exposed
  • Input is validated and sanitised before use
  • Dependencies added are not known to be vulnerable
  • Authentication / authorisation checks are in place where required

Performance

  • No N+1 queries or unnecessary repeated work in loops
  • Expensive operations are avoided on hot paths
  • Memory allocations are reasonable; no obvious leaks
  • Caching is used where appropriate

Readability & maintainability

  • Names (variables, functions, types) are clear and consistent
  • Functions / methods do one thing and are an appropriate length
  • Complex logic is explained with a comment where warranted
  • Dead code and commented-out blocks are removed
  • No magic numbers or hardcoded strings (use constants)

Tests

  • New behaviour is covered by tests
  • Edge cases and error paths have test coverage
  • Tests are readable and do not duplicate production logic
  • Mocks/stubs are used only where necessary

Style & conventions

  • Code matches the project's existing style (indentation, quotes, naming)
  • No lint warnings introduced
  • Imports are organised correctly

4. Output format

Produce the review in this exact structure:

## Code Review — [file or PR title]
**Reviewed:** [date]
**Diff size:** ~[N] lines changed

---

### 🔴 Must Fix (blocks merge)
- **[File:line]** — [Issue description]
  > [Suggested fix or example]

### 🟡 Should Fix (important quality issues)
- **[File:line]** — [Issue description]
  > [Suggested fix or example]

### 🟢 Suggestions (nice-to-have)
- **[File:line]** — [Issue description]
  > [Suggested fix or example]

### ✅ Looks Good
- [List of things done well]

---

### Summary
[2-3 sentence overall assessment. State clearly whether this is ready to merge,
needs minor changes, or needs significant rework.]

Rules

  • Be specific: always include file name and line number.
  • Show a concrete fix, not just a description of the problem.
  • Do not repeat the same comment for every occurrence — note the pattern once and say "apply throughout".
  • Praise genuinely good choices to reinforce good habits.
  • Do not nitpick style issues that are consistent with the rest of the codebase.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

36.94%
按下载量换算31

Claude

28.46%
按下载量换算24

Cursor

18.61%
按下载量换算16

Gemini CLI

9.44%
按下载量换算8

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills