代码审查与质量检测:检查代码规范、潜在Bug、性能问题、安全漏洞,输出审查报告与改进建议。Invoke when user asks 代码审查、Code Review、代码质量、代码检查、代码优化.
编程
Code Review Checklist
试用代码审查清单 - A comprehensive code review checklist and guidance tool. Use when user asks about 代码审查、代码检查、PR review、代码质量, or wants to conduct or prepare for a cod...
它能做什么
This skill provides a systematic approach to code reviews. It offers comprehensive checklist items across multiple dimensions of code quality, helps reviewers focus on high-impact areas, and guides developers in preparing code for review. Designed to make code reviews more efficient and thorough.
技能文档
Code Review Checklist (代码审查清单)
Overview
This skill provides a systematic approach to code reviews. It offers comprehensive checklist items across multiple dimensions of code quality, helps reviewers focus on high-impact areas, and guides developers in preparing code for review. Designed to make code reviews more efficient and thorough.
When to Use This Skill
- Preparing code for pull request review
- Conducting a code review as a reviewer
- Self-reviewing own code before submission
- Establishing code review standards for a team
- Training new developers on review best practices
- Auditing code quality in a codebase
What This Skill Provides
1. Predefined Checklists
Comprehensive checklist items organized by category:
- Code correctness and logic
- Code style and readability
- Performance and efficiency
- Security considerations
- Error handling
- Testing coverage
- Documentation
- Architecture and design patterns
2. Review Guidance
- What to look for in each category
- Red flags and common issues
- Best practices specific to language/framework
- Questions to ask the author
3. Review Workflow
- Systematic approach to reviewing
- Priority ordering of checks
- Time allocation guidance
- Documentation requirements
Checklist Categories
1. Correctness & Logic
- Code produces expected output
- Edge cases are handled
- No off-by-one errors
- Logic is sound and complete
- No infinite loops or recursion issues
- Proper use of data structures
2. Code Style & Readability
- Follows project coding standards
- Naming is clear and descriptive
- Functions are appropriately sized
- Code is not duplicated (DRY principle)
- Complex logic has comments
- Formatting is consistent
3. Performance & Efficiency
- No unnecessary loops or iterations
- Proper use of caching when applicable
- Database queries are optimized
- No memory leaks
- Appropriate algorithmic complexity
- Resources are properly released
4. Security
- Input validation on all user inputs
- No SQL injection vulnerabilities
- No XSS vulnerabilities
- Secrets not hardcoded
- Proper authentication/authorization
- Sensitive data properly protected
- No security misconfigurations
5. Error Handling
- Errors are caught and handled appropriately
- Error messages are user-friendly
- No empty catch blocks
- Logging is appropriate
- Graceful degradation where needed
- No exposing internal error details
6. Testing
- Unit tests exist for new code
- Tests cover happy path and edge cases
- Tests are maintainable
- Mock usage is appropriate
- Test coverage meets requirements
- No flaky tests introduced
7. Documentation
- Public APIs are documented
- Complex logic has comments
- README updated if needed
- API changes are documented
- Breaking changes are noted
8. Architecture & Design
- Follows project architecture patterns
- Single Responsibility Principle followed
- Dependencies are properly injected
- Coupling is minimized
- Changes are localized appropriately
- No tech debt introduced unnecessarily
Language-Specific Considerations
JavaScript/TypeScript
- Proper async/await usage
- TypeScript types are correct
- No 'any' type abuse
- ESLint rules followed
Python
- PEP 8 compliance
- Type hints where appropriate
- Docstrings for public functions
- No deprecated imports
Java
- Null safety considerations
- Resource management (try-with-resources)
- Stream API usage
- Concurrent access considerations
Go
- Error handling conventions
- Goroutine leak prevention
- Context usage
- Naming conventions
Review Workflow
Step 1: Context (2-3 min)
- Read PR description and motivation
- Understand what changed and why
- Check related issues or docs
Step 2: Overview (3-5 min)
- Scan changed files
- Identify high-risk areas
- Note files needing deep review
Step 3: Detailed Review (15-30 min)
- Follow checklist by priority
- Comment on issues found
- Ask clarifying questions
- Suggest improvements
Step 4: Summary (3-5 min)
- Summarize findings
- Categorize issues (Blocking/Suggestion/Question)
- Approve or request changes
Usage Examples
As a Reviewer
"用代码审查清单检查这个PR"
"帮我审查这个函数的逻辑"
"检查这段代码有没有安全问题"
"看看这个文件有哪些可以改进的地方"
As a Developer
"帮我准备代码审查"
"自审查这份代码,有什么遗漏?"
"检查这段代码的测试覆盖"
"这个代码符合项目规范吗?"
For Team Standards
"生成一个代码审查检查清单"
"我们团队的代码审查标准是什么?"
"前端代码审查有什么特殊要求?"
Output Format
For each review, output:
## Code Review: [PR/Change Title]
### Summary
- Files changed: X
- Lines added/removed: +X/-X
- Risk level: [Low/Medium/High]
### Findings
#### 🔴 Blocking Issues
- [Issue description] - [File:Line] - [Suggestion]
#### 🟡 Suggestions
- [Suggestion] - [File:Line]
#### 🟢 Good Practices Noted
- [Positive observation]
### Checklist Status
- [x] Correctness
- [x] Style
- [ ] Security (needs work)
- [x] Performance
### Recommendation
[Approve / Request Changes / Discuss]
### Action Items
- [ ] Item 1
- [ ] Item 2
Integration with Development Workflow
This skill integrates with:
github— For reviewing PRs directlycoding-agent— For automated code quality checksopencli— For running linters and formatters
Limitations
- Cannot execute code to verify correctness
- Cannot know full system context
- Best practices may vary by project
- Language-specific items may be incomplete for niche languages
Acceptance Criteria
- ✓ Provides comprehensive checklist coverage
- ✓ Can customize for different languages/frameworks
- ✓ Identifies common issues efficiently
- ✓ Helps categorize issue severity
- ✓ Provides actionable feedback
- ✓ Saves time in review process
- ✓ Helps developers learn and improve
相关技能
Review code against Android/iOS/TypeScript/Go/general rules. Triggers: review, code review, check my changes, 帮我看看代码, commit hash, PR URL. Read-only, never modifies code.
Generate HTML code review pages with risk tags, diff highlights, and file-level annotations. 当用户需要代码审查可视化、PR审查报告、代码diff高亮、风险标签标注、审查页面生成时使用。
Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensi Use when 需要Development领域自动化处理、数据分析和流程编排时使用。不适用于无明确需求的模糊场景。
拿到资深工程师风格的代码审查:严重度分级、OWASP 安全审计与可直接粘贴的 PR 评语
コードレビューを実施する際に使用。プルリクエストのレビュー、コード品質チェック、 セキュリティ検査、パフォーマンス分析を行う。「レビューして」「コードをチェック」 「品質を確認」などの指示で起動。
haidong 的更多技能
浏览全部技能Generate a Table of Contents (TOC) from Markdown headings. Supports heading extraction, multiple output formats, and configurable level filtering.
Interpret and explain lines from the Dao De Jing (道德经) in plain Chinese. Use when the user quotes a sentence, a chapter, or a phrase from the Dao De Jing, as...