정보
이 스킬은 개발자가 체계적인 네이밍, 작은 함수, 깔끔한 오류 처리를 통해 코드 가독성을 작성하고 개선하는 데 도움을 줍니다. 코드 스멜, 긴 함수, 네이밍 규칙을 다루거나 SRP, 주석, 포맷팅에 대한 코드 리뷰 시 사용하세요. 이 스킬은 유지보수 가능한 코드 품질에 중점을 두며, 아키텍처나 구체적인 리팩토링 작업은 다른 관련 스킬로 안내합니다.
빠른 설치
Claude Code
추천npx skills add wondelai/skills -a claude-code/plugin add https://github.com/wondelai/skillsgit clone https://github.com/wondelai/skills.git ~/.claude/skills/clean-codeClaude Code에서 이 명령을 복사하여 붙여넣어 스킬을 설치하세요
문서
Clean Code Framework
A disciplined approach to writing code that communicates intent, minimizes surprises, and welcomes change. Apply these principles when writing new code, reviewing pull requests, refactoring legacy systems, or advising on code quality.
Core Principle
Code is read far more often than it is written — optimize for the reader. The read-to-write ratio is well over 10:1, so every naming choice, function boundary, and formatting decision either adds clarity or adds cost. Clean code reads like well-written prose: names reveal intent, functions tell a story one step at a time, and the Boy Scout Rule applies — always leave the code cleaner than you found it.
Scoring
Goal: 10/10. Rate any code 0-10 against the principles below. Report the current score and the specific improvements needed to reach 10/10.
- 9-10: Names reveal intent, functions are small and focused, error handling is consistent, tests are clean and comprehensive
- 7-8: Mostly clean with minor naming ambiguities or a few long functions; tests may lack edge cases
- 5-6: Mixed — good patterns alongside unclear names, duplicated logic, or inconsistent error handling
- 3-4: Long multi-purpose functions, misleading names, poor or missing tests
- 1-2: Nearly unreadable — magic numbers, cryptic abbreviations, no structure, no tests
The Clean Code Framework
Six disciplines for writing code that communicates clearly and adapts to change:
1. Meaningful Names
Core concept: Names should reveal intent, avoid disinformation, and make the code read like prose. If a name requires a comment to explain it, the name is wrong.
Why it works: Names are the most pervasive form of documentation — a well-chosen name eliminates the need to read the implementation; a poor one forces every reader to reverse-engineer intent.
Key insights:
- A name should answer why it exists, what it does, and how it is used
- No encodings, prefixes, or type information (no Hungarian notation); single letters only for tiny-scope loop counters
- Classes are nouns; methods are verbs
- One word per concept: don't mix
fetch,retrieve, andget - Longer scope demands a longer, more descriptive name
- Rename freely — IDEs make it trivial
Code applications:
| Context | Pattern | Example |
|---|---|---|
| Variables | Intention-revealing | elapsedTimeInDays not d |
| Booleans | Predicate phrasing | isActive, hasPermission, canEdit |
| Functions | Verb + noun | calculateMonthlyRevenue() not calc() |
| Classes | Noun naming the responsibility | InvoiceGenerator not InvoiceManager |
See references/naming-conventions.md when renaming or reviewing names — per-language conventions, pronounceable/searchable tables, and before/after examples.
2. Functions
Core concept: Functions should be small, do one thing, and do it well — ideally 4-6 lines, zero to two arguments, one level of abstraction.
Why it works: Small single-purpose functions are easy to name, understand, test, and reuse; long functions hide bugs, resist testing, and accumulate responsibilities.
Key insights:
- Step-Down Rule: code reads top-down, each function calling the next level of abstraction
- Argument count: zero best, one fine, two acceptable, three+ requires justification
- Flag arguments are a smell — the function does two things; split it
- Command-Query Separation: change state or return a value, never both
- Extract till you drop: if you can pull out a named function, do it
- No hidden side effects — the name must tell the whole truth
Code applications:
| Context | Pattern | Example |
|---|---|---|
| Long function | Extract named steps | validateInput(); transformData(); saveRecord(); |
| Flag argument | Split into two functions | renderForPrint() / renderForScreen() not render(isPrint) |
| Error cases | Guard clauses at top | Early return for errors, single happy path |
| Many arguments | Introduce parameter object | new DateRange(start, end) not report(start, end, format, locale) |
| Side effects | Make effects explicit | checkPassword() that starts a session → rename or separate |
See references/functions-and-methods.md when splitting a long function — argument-count rules, command-query separation, and step-down worked examples.
3. Comments and Formatting
Core concept: A comment is a failure to express yourself in code. When comments are necessary, they explain why, never what. Formatting creates the visual structure that makes code scannable.
Why it works: Comments rot — code changes but comments often don't, creating documentation worse than none. Clean formatting lets developers scan code like a newspaper: headlines first, details on demand.
Key insights:
- The best comment is a well-named extracted function
- Acceptable: legal headers, TODOs, public API docs, genuine "why" explanations
- Commented-out code and journal comments: delete — version control remembers
- Vertical openness between concepts; vertical density within them; declare variables near usage
- Newspaper metaphor: high-level functions at the top of the file, details below
Code applications:
| Context | Pattern | Example |
|---|---|---|
| Explaining "what" | Replace with better name | // check if eligible → isEligible() |
| Explaining "why" | Keep as comment | // RFC 7231 requires this header for proxies |
| Commented-out code | Delete it | Trust version control |
| Team formatting | Decide once, automate | Prettier, Black, gofmt |
See references/comments-formatting.md when deciding whether a comment earns its place — good-vs-bad comment catalog and vertical-formatting rules.
4. Error Handling
Core concept: Error handling is a separate concern from business logic. Use exceptions rather than return codes, provide context with every exception, and never return or pass null.
Why it works: Return codes clutter the happy path with checks; exceptions separate the two cleanly. Returning null forces null checks on every caller, and one missing check crashes far from the source.
Key insights:
- Write the try-catch first — it defines a transaction boundary
- Prefer unchecked exceptions — checked ones violate the Open/Closed Principle
- Define exception classes by the caller's needs, not the failure type
- Don't return null (use empty collections, Optional, or throw); don't pass null either
- Special Case / Null Object pattern: return an object with default behavior instead of null
Code applications:
| Context | Pattern | Example |
|---|---|---|
| Null returns | Empty collection or Optional | return Collections.emptyList() not return null |
| Error codes | Replace with exceptions | throw new InsufficientFundsException(balance, amount) |
| Third-party APIs | Wrap with adapter | PortfolioService wraps the vendor API, translates its exceptions |
| Special cases | Null Object pattern | GuestUser with default behavior instead of null checks |
| Context in errors | Include operation + state | "Failed to save invoice #1234 for customer 'Acme'" |
See references/error-handling.md when designing exception or null strategy — Special Case pattern and third-party-API wrapping examples.
5. Unit Testing
Core concept: Tests are first-class code, kept clean with the same discipline as production code. Dirty tests are worse than no tests — they become a liability that slows every change.
Why it works: Clean tests are executable documentation and a safety net for refactoring; dirty tests make every modification a fight through incomprehensible test code.
Key insights:
- Three Laws of TDD: write a failing test first; only enough test to fail; only enough code to pass
- One concept per test — one logical assertion, not necessarily one assert
- F.I.R.S.T.: Fast, Independent, Repeatable, Self-validating, Timely
- Build a domain-specific testing language: helpers that read like a DSL
- Refactor test code as readily as production code
Code applications:
| Context | Pattern | Example |
|---|---|---|
| Test structure | Arrange-Act-Assert | Setup, execute, verify — clearly separated |
| Test naming | Scenario + expected behavior | shouldRejectExpiredToken not test1 |
| Shared setup | Builder/factory helpers | aUser().withRole(ADMIN).build() |
| Flaky tests | Remove external dependencies | Mock time, network, file system |
See references/testing-principles.md when writing or cleaning tests — TDD laws, F.I.R.S.T. expanded, and clean-test patterns.
6. Code Smells and Heuristics
Core concept: Smells are surface indicators of deeper design problems — learn to recognize them quickly and apply targeted refactorings instead of vague "cleanup".
Why it works: Smells are heuristics that point toward likely problems without deep analysis, turning code review instinct into specific, repeatable moves.
Key insights:
- Function smells: too many arguments, output arguments, flag arguments, dead functions
- General smells: duplication, wrong level of abstraction, feature envy, magic numbers
- Test smells: insufficient coverage, skipped tests, untested boundary conditions and failure paths
- Refactor in small, tested steps — never refactor and add features simultaneously
- Boy Scout Rule: leave the code cleaner than you found it
Code applications:
| Context | Pattern | Example |
|---|---|---|
| Duplication | Extract shared logic | Common validation → validateEmail() helper |
| Feature envy | Move method to the data's class | order.calculateTotal() not calculator.total(order) |
| Dead code | Delete it | Remove unused functions, unreachable branches |
| Magic numbers | Named constants | MAX_LOGIN_ATTEMPTS = 5 not bare 5 |
| Shotgun surgery | Consolidate related changes | Group scattered logic into a single module |
See references/code-smells.md when a smell is hard to name — the full catalog by category, each paired with its targeted refactoring.
Common Mistakes
| Mistake | Why It Fails | Fix |
|---|---|---|
| Abbreviating names | Saves seconds writing, costs hours reading | Full descriptive names; IDEs autocomplete |
| "Clever" one-liners | Impressive to write, impossible to debug | Expand into readable named steps |
| Comments instead of refactoring | Comments rot; code is the truth | Extract a well-named function instead |
| Catching generic exceptions | Swallows bugs along with expected errors | Catch specific exceptions; let the rest propagate |
| No tests for error paths | Happy path works, edge cases crash | Test every branch, boundary, and failure mode |
| Premature optimization | Obscures intent for marginal gains | Clean first; optimize measured bottlenecks |
| God classes | One 2000-line class does everything | Apply SRP — split by responsibility |
| Refactoring without tests | No safety net for regressions | Write characterization tests first |
| Inconsistent conventions | Every file feels like a different codebase | Agree on style; enforce with linters and formatters |
| Returning null everywhere | Null checks spread like a virus | Optional, empty collections, or Null Object |
Quick Diagnostic
| Question | If No | Action |
|---|---|---|
| Can you understand each function without reading its body? | Names don't reveal intent | Rename to describe what it does |
| Are all functions under 20 lines? | Functions do too many things | Extract sub-operations into named helpers |
| Zero commented-out code blocks? | Dead code creating confusion | Delete — version control has history |
| Is error handling separate from business logic? | Try-catch clutters the main flow | Extract handlers; exceptions over return codes |
| Does every class have a single responsibility? | Classes accumulate unrelated duties | Split into focused, well-named classes |
| Is there a test for every public method? | No safety net for changes | Add tests before changing further |
| Are test names descriptive of behavior? | Failures are hard to interpret | Rename to shouldDoXWhenY |
| Is duplication below 3 occurrences? | Copy-paste spreading bugs | Extract shared logic (§6) |
| Are magic numbers named constants? | Intent hidden behind raw values | Name the constant (§6) |
| Do all tests run in under 10 seconds? | Slow tests don't get run | Mock external deps; split integration tests |
Further Reading
Based on Robert C. Martin's seminal guide to software craftsmanship:
- "Clean Code: A Handbook of Agile Software Craftsmanship" by Robert C. Martin
- "The Clean Coder: A Code of Conduct for Professional Programmers" by Robert C. Martin
- "Clean Architecture: A Craftsman's Guide to Software Structure and Design" by Robert C. Martin
- "Refactoring: Improving the Design of Existing Code" by Martin Fowler
About the Author
Robert C. Martin ("Uncle Bob") has been programming since 1970, co-authored the Agile Manifesto, and founded Uncle Bob Consulting and Clean Coders. His books — Clean Code, The Clean Coder, Clean Architecture, and Clean Agile — shaped how a generation of developers think about code quality, and his core stance is that the only way to go fast is to go well.
GitHub 저장소
자주 묻는 질문
clean-code Skill이란 무엇인가요?
clean-code은(는) wondelai이(가) 만든 Claude Skill입니다. Skill은 Claude가 필요할 때 불러오는 지침과 리소스를 묶어 추가 프롬프트 없이 clean-code 관련 작업을 수행할 수 있게 합니다.
clean-code은(는) 어떻게 설치하나요?
이 페이지의 설치 명령을 사용하세요. clean-code을(를) Claude Code 플러그인으로 추가하거나 저장소를 skills 디렉터리에 복제한 다음 Claude를 다시 시작해 Skill을 불러옵니다.
clean-code은(는) 어떤 카테고리에 속하나요?
clean-code은(는) 테스팅 카테고리에 속합니다.
clean-code은(는) 무료로 사용할 수 있나요?
네. clean-code은(는) AIMCP에 등록되어 있으며 무료로 설치할 수 있습니다.
연관 스킬
이 Claude Skill은 MMLU, GSM8K를 포함한 60개 이상의 표준화된 학술 과제에서 LLM 성능을 벤치마크하기 위해 lm-evaluation-harness를 실행합니다. 개발자들이 모델 품질을 비교하고, 학습 진행 상황을 추적하거나 학술 결과를 보고할 수 있도록 설계되었습니다. 이 도구는 HuggingFace와 vLLM 모델을 포함한 다양한 백엔드를 지원합니다.
이 스킬은 cron 표현식을 사용하여 Worker를 스케줄링하기 위한 Cloudflare Cron Triggers 구현에 관한 포괄적인 지식을 제공합니다. 주기적 작업, 유지보수 작업, 자동화된 워크플로우 설정 방법을 다루며, 잘못된 cron 표현식이나 시간대 문제 같은 일반적인 이슈들을 해결하는 방법을 포함합니다. 개발자들은 이를 통해 스케줄된 핸들러 구성, cron 트리거 테스트, Workflows 및 Green Compute와의 연동 작업을 수행할 수 있습니다.
이 Claude Skill은 Python 스크립트를 통해 로컬 웹 애플리케이션을 테스트하기 위한 Playwright 기반 툴킷을 제공합니다. 프론트엔드 검증, UI 디버깅, 스크린샷 캡처, 로그 확인 기능을 지원하며 서버 라이프사이클을 관리합니다. 브라우저 자동화 작업에 사용하되 컨텍스트 오염을 방지하기 위해 소스 코드를 읽지 않고 스크립트를 직접 실행하세요.
이 스킬은 테스트 통과를 확인한 후 체계적인 통합 옵션을 제시하여 개발자가 완성된 작업을 마무리하도록 돕습니다. 구현이 완료된 후 머지, PR 생성, 브랜치 정리와 같은 워크플로우를 안내합니다. 코드가 준비되고 테스트가 완료되었을 때 개발 프로세스를 체계적으로 마무리하기 위해 사용하세요.
