Token导航 LogoToken导航TokenDH.com
开发只读github未标认证来源可访问clear审计通过

code-reviewer代码审查员

Agent Skill

code-reviewer 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

2,514

周安装

108

GitHub Stars

76

下载量

881
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

3

许可证

MIT

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-reviewer(代码审查员)
来源仓库:https://github.com/404kidwiz/claude-supercode-skills
仓库路径:skills/code-reviewer
安装命令:
npx skills add https://github.com/404kidwiz/claude-supercode-skills --skill code-reviewer
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/404kidwiz/claude-supercode-skills --skill code-reviewer

简介

code-reviewer 提供代码审查专长,聚焦正确性、安全性、性能和可维护性。

  • 适用于审查拉取请求、执行安全审计、识别合并前缺陷及验证最佳实践。
  • 在 Codex、Claude、Cursor、Gemini CLI 中调用,需结合仓库状态或协作事项使用。
  • 安装命令为 npx skills add https://github.com/404kidwiz/claude-supercode-skills --skill code-reviewer。
  • 建议确认权限范围、维护状态,避免触发联网或文件读写等敏感操作。

SKILL.md

Code Reviewer

Purpose

Provides thorough code review expertise with focus on correctness, security, performance, and maintainability. Identifies bugs, security vulnerabilities, and code quality issues while suggesting improvements.

When to Use

  • Reviewing pull requests or code changes
  • Performing security audits on code
  • Identifying potential bugs before merge
  • Ensuring code follows best practices
  • Checking for performance issues
  • Validating error handling
  • Reviewing architectural decisions in code

Quick Start

Invoke this skill when:

  • Reviewing pull requests or code changes
  • Performing security audits on code
  • Identifying potential bugs before merge
  • Ensuring code follows best practices
  • Checking for performance issues

Do NOT invoke when:

  • Debugging runtime issues (use debugger)
  • Refactoring code structure (use refactoring-specialist)
  • Writing new code (use language-specific skills)
  • Reviewing system architecture (use architect-reviewer)

Decision Framework

Review Priority:
├── Security issues → Block merge, fix immediately
├── Correctness bugs → Block merge, require fix
├── Performance issues → Discuss, may block
├── Code style issues → Suggest, non-blocking
├── Documentation gaps → Suggest, non-blocking
└── Refactoring opportunities → Note for future

Core Workflows

1. Pull Request Review

  1. Understand the intent from PR description
  2. Review for correctness and logic errors
  3. Check for security vulnerabilities
  4. Assess performance implications
  5. Verify error handling completeness
  6. Check test coverage
  7. Provide actionable feedback

2. Security-Focused Review

  1. Check input validation and sanitization
  2. Review authentication and authorization
  3. Look for injection vulnerabilities
  4. Verify sensitive data handling
  5. Check for hardcoded secrets
  6. Review dependency security
  7. Assess cryptographic usage

3. Performance Review

  1. Identify N+1 query patterns
  2. Check for unnecessary allocations
  3. Review algorithm complexity
  4. Assess caching opportunities
  5. Check for blocking operations
  6. Review database query efficiency

Best Practices

  • Review code, not the author
  • Be specific about issues and fixes
  • Explain the "why" behind suggestions
  • Prioritize comments by severity
  • Acknowledge good patterns too
  • Use automated tools first (linters, SAST)

Anti-Patterns

Anti-PatternProblemCorrect Approach
Nitpicking styleWastes time, frustrates authorsUse automated formatters
No contextReviewer doesn't understand changesRead PR description, linked issues
Blocking on opinionsDelays delivery unnecessarilyDistinguish must-fix from nice-to-have
Drive-by reviewsComments without resolutionFollow through on discussions
No positive feedbackDemoralizing for authorsHighlight good patterns

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

04

需要参考平台分布和安装热度时

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

补充不同宿主或平台的使用分布数据

能力 5

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

Claude Code

27.68%
按下载量换算244

OpenCode

24.86%
按下载量换算219

Codex

18.35%
按下载量换算162

Gemini CLI

11.3%
按下载量换算100

Antigravity

7.44%
按下载量换算66

Cursor

3.16%
按下载量换算28

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。

来源信息

继续浏览同类 Skills