Token导航 LogoToken导航TokenDH.com
研究检索敏感数据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 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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

465

周安装

19

GitHub Stars

8

下载量

149
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/vamseeachanta/workspace-hub
仓库路径:skills/code-reviewer
安装命令:
npx skills add https://github.com/vamseeachanta/workspace-hub --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/vamseeachanta/workspace-hub --skill code-reviewer

简介

code-reviewer 用于查找、检索和筛选相关信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词或任务场景快速定位候选结果。
  • 通过 npx skills add 命令从指定仓库安装,需结合原始 README 确认具体用法。
  • 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。
  • 当前暂无更多细节,建议参考来源仓库了解具体实现方式。

SKILL.md

Code Reviewer Skill

Overview

This skill provides comprehensive code review capabilities across multiple programming languages including TypeScript, JavaScript, Python, Swift, and Kotlin. Focus areas: code quality, security, performance, and maintainability.

Quick Start

  1. Identify scope - Files or PR to review
  2. Run analysis - Check code quality, security, performance
  3. Document findings - Categorize by severity
  4. Provide recommendations - Actionable improvements
  5. Generate report - Structured review output

When to Use

  • Pull request reviews
  • Pre-merge quality gates
  • Code audit requirements
  • Security assessments
  • Performance optimization reviews
  • Onboarding code familiarization

Review Categories

1. Code Quality

AspectCheck For
ClarityReadable, self-documenting code
NamingDescriptive, consistent conventions
StructureSingle responsibility, appropriate abstraction
DRYNo unnecessary duplication
ComplexityCyclomatic complexity within limits

2. Security

AspectCheck For
Input validationAll user inputs validated
SQL injectionParameterized queries
XSSOutput encoding
Auth/AuthZProper authentication and authorization
SecretsNo hardcoded credentials
DependenciesUpdated, no known vulnerabilities

3. Performance

AspectCheck For
AlgorithmsAppropriate time complexity
MemoryNo leaks, efficient usage
DatabaseOptimized queries, proper indexing
CachingAppropriate cache usage
AsyncNon-blocking operations where needed

4. Testing

AspectCheck For
CoverageCritical paths tested
QualityMeaningful assertions
IsolationTests don't depend on each other
Edge casesBoundary conditions covered
MockingMinimal, appropriate mocking

Review Process

Step 1: Context Gathering

## Review Context
- **PR/Files:** [identifier]
- **Author:** [name]
- **Purpose:** [feature/bugfix/refactor]
- **Related:** [issues/tickets]

Step 2: High-Level Analysis

Scan for:

  • Overall change scope
  • Architectural impact
  • Breaking changes
  • New dependencies

Step 3: Detailed Review

For each file:

### [filename]

**Changes:** [summary]

**Findings:**
- [severity] [category]: [description]
  - Location: line X
  - Recommendation: [action]

Step 4: Security Deep Dive

Special attention to:

  • Authentication flows
  • Data handling
  • API endpoints
  • Configuration files
  • Environment variables

Step 5: Test Verification

  • Run existing tests
  • Verify new test coverage
  • Check test quality
  • Identify missing tests

Severity Levels

LevelDescriptionAction
CriticalSecurity vulnerability, data loss riskBlock merge
HighBugs, significant issuesMust fix
MediumCode quality, maintainabilityShould fix
LowStyle, minor improvementsConsider
InfoSuggestions, observationsOptional

Review Checklist

General

  • Code compiles/builds without errors
  • No merge conflicts
  • Branch is up to date with target
  • Commit messages are clear

Functionality

  • Requirements are met
  • Edge cases handled
  • Error handling is appropriate
  • No regression in existing functionality

Code Quality

  • Follows project coding standards
  • No code smells or anti-patterns
  • Appropriate comments (not excessive)
  • No dead code or debug statements

Security

  • No hardcoded secrets
  • Input validation present
  • Output encoding where needed
  • Proper error messages (no info leak)

Testing

  • New code has tests
  • All tests pass
  • Test coverage adequate
  • Tests are meaningful

Documentation

  • README updated if needed
  • API docs updated if applicable
  • Breaking changes documented

Common Anti-Patterns

Code Smells

PatternProblemSolution
God ObjectClass does too muchSplit responsibilities
Feature EnvyMethod uses other class moreMove to appropriate class
Long MethodHard to understandExtract smaller methods
Magic NumbersUnclear meaningUse named constants
Deep NestingHard to followEarly returns, extraction

Security Issues

PatternProblemSolution
SQL ConcatInjection riskParameterized queries
Eval UsageCode injectionSafe alternatives
Weak CryptoBreakable encryptionStrong algorithms
CORS *Access control bypassSpecific origins
Console SecretsCredential exposureRemove before commit

Review Report Template

# Code Review Report

## Summary
- **Reviewed:** [files/PR]
- **Date:** [date]
- **Reviewer:** [name]
- **Overall:** [APPROVE/REQUEST_CHANGES/COMMENT]

## Statistics
- Files reviewed: X
- Lines changed: +Y/-Z
- Critical issues: N
- High issues: N
- Medium issues: N

## Critical/High Findings

### [Finding Title]
- **Severity:** Critical/High
- **Category:** Security/Bug/Performance
- **Location:** file:line
- **Description:** [details]
- **Recommendation:** [action]
- **Code suggestion:**
  \`\`\`language
  // suggested fix
  \`\`\`

## Medium/Low Findings
[Grouped by category]

## Positive Observations
- [Good practices noticed]

## Recommendations
1. [Priority improvement]
2. [Secondary improvement]

## Test Coverage
- Current: X%
- Critical paths: Y%
- Recommendation: [action]

Best Practices

Do

  1. Review in small batches (200-400 lines ideal)
  2. Focus on logic, not style (use linters)
  3. Ask questions rather than demand changes
  4. Acknowledge good code
  5. Provide specific, actionable feedback
  6. Test the changes locally when possible

Don't

  1. Nitpick style issues
  2. Rewrite author's code in comments
  3. Leave vague feedback ("this is wrong")
  4. Review when fatigued
  5. Approve without understanding
  6. Block for preferences, not issues

Error Handling

SituationAction
Too large PRRequest split into smaller PRs
Missing contextAsk author for explanation
Unclear requirementsDefer to requirements review
DisagreementEscalate with evidence

Metrics

MetricTargetDescription
Review turnaround<24hTime to first review
Defect detection>80%Issues caught before merge
False positive rate<10%Unnecessary comments
Review thoroughness100%All critical areas covered

Technology-Specific Notes

TypeScript/JavaScript

  • Check for any/unknown abuse
  • Verify type safety
  • Review async/await handling
  • Check for memory leaks in React

Python

  • PEP 8 compliance
  • Type hints present
  • Exception handling
  • Virtual environment usage

Go

  • Error handling patterns
  • Goroutine leaks
  • Interface usage
  • Package organization

Related Skills


Version History

  • 1.0.0 (2026-01-19): Initial release adapted from alirezarezvani/claude-skills

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

29.97%
按下载量换算45

windsurf

22.6%
按下载量换算34

trae

19.09%
按下载量换算28

OpenCode

13.36%
按下载量换算20

Cursor

7.99%
按下载量换算12

Codex

3.82%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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