Systematic code refactoring based on Martin Fowler's methodology. Use when users ask to refactor code, improve code structure, reduce technical debt, clean up legacy code, eliminate code smells, or improve code maintainability. This skill guides through a phased approach with research, planning, and safe incremental implementation.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
A systematic approach to refactoring code based on Martin Fowler's Refactoring: Improving the Design of Existing Code (2nd Edition). This skill emphasizes safe, incremental changes backed by tests.
"Refactoring is the process of changing a software system in such a way that it does not alter the external behavior of the code yet improves its internal structure." — Martin Fowler
Phase 1: Research & Analysis
↓
Phase 2: Test Coverage Assessment
↓
Phase 3: Code Smell Identification
↓
Phase 4: Refactoring Plan Creation
↓
Phase 5: Incremental Implementation
↓
Phase 6: Review & Iteration
Before starting, clarify:
Present findings to user:
"Refactoring without tests is like driving without a seatbelt." — Martin Fowler
Tests are the key enabler of safe refactoring. Without them, you risk introducing bugs.
Check for existing tests
# Look for test files
find . -name "*test*" -o -name "*spec*" | head -20
Run existing tests
# JavaScript/TypeScript
npm test
# Python
pytest -v
# Java
mvn test
Check coverage (if available)
# JavaScript
npm run test:coverage
# Python
pytest --cov=.
If tests exist and pass:
If tests are missing or incomplete: Present options:
If tests are failing:
For each function being refactored, ensure tests cover:
Use the "red-green-refactor" cycle:
Symptoms of deeper problems in code. They're not bugs, but indicators that the code could be improved.
See references/code-smells.md for the complete catalog.
| Smell | Signs | Impact |
|---|---|---|
| Long Method | Methods > 30-50 lines | Hard to understand, test, maintain |
| Duplicated Code | Same logic in multiple places | Bug fixes needed in multiple places |
| Large Class | Class with too many responsibilities | Violates Single Responsibility |
| Feature Envy | Method uses another class's data more | Poor encapsulation |
| Primitive Obsession | Overuse of primitives instead of objects | Missing domain concepts |
| Long Parameter List | Methods with 4+ parameters | Hard to call correctly |
| Data Clumps | Same data items appearing together | Missing abstraction |
| Switch Statements | Complex switch/if-else chains | Hard to extend |
| Speculative Generality | Code "just in case" | Unnecessary complexity |
| Dead Code | Unused code | Confusion, maintenance burden |
Automated Analysis (if scripts available)
python scripts/detect-smells.py <file>
Manual Review
Prioritization Focus on smells that:
Present to user:
For each smell, select an appropriate refactoring from the catalog.
See references/refactoring-catalog.md for the complete list.
| Code Smell | Recommended Refactoring(s) |
|---|---|
| Long Method | Extract Method, Replace Temp with Query |
| Duplicated Code | Extract Method, Pull Up Method, Form Template Method |
| Large Class | Extract Class, Extract Subclass |
| Feature Envy | Move Method, Move Field |
| Primitive Obsession | Replace Primitive with Object, Replace Type Code with Class |
| Long Parameter List | Introduce Parameter Object, Preserve Whole Object |
| Data Clumps | Extract Class, Introduce Parameter Object |
| Switch Statements | Replace Conditional with Polymorphism |
| Speculative Generality | Collapse Hierarchy, Inline Class, Remove Dead Code |
| Dead Code | Remove Dead Code |
Use the template at templates/refactoring-plan.md.
For each refactoring:
CRITICAL: Introduce refactoring gradually in phases.
Phase A: Quick Wins (Low risk, high value)
Phase B: Structural Improvements (Medium risk)
Phase C: Architectural Changes (Higher risk)
Before implementation:
"Change → Test → Green? → Commit → Next step"
For each refactoring step:
Pre-check
Make ONE small change
Verify
If tests pass (green)
If tests fail (red)
Each commit should be:
Example commit messages:
refactor: Extract calculateTotal() from processOrder()
refactor: Rename 'x' to 'customerCount' for clarity
refactor: Remove unused validateOldFormat() method
After each sub-phase, report to user:
Run complexity analysis before and after:
python scripts/analyze-complexity.py <file>
Present improvements:
Present final results:
Discuss with user:
Always pause and consult user when:
Before:
function processOrder(order) {
// 150 lines of code with:
// - Duplicated validation logic
// - Inline calculations
// - Mixed responsibilities
}
Refactoring Steps:
After:
function processOrder(order) {
validateOrder(order);
const total = calculateOrderTotal(order);
notifyCustomer(order, total);
return { order, total };
}
scripts/analyze-complexity.py - Analyze code complexity metricsscripts/detect-smells.py - Automated smell detectionLast Updated: August 4, 2026 Claude Code Version: 2.1.220 Sources:
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Draft a blog post from ideas and resources. Use when users want to write a blog post, create content from research, or draft articles. Guides through research, brainstorming, outlining, and iterative drafting with version control.
日本語の概要は準備中です。原文の説明を表示しています。
Soạn thảo bài đăng blog từ ý tưởng và tài nguyên. Sử dụng khi người dùng muốn viết bài blog, tạo nội dung từ nghiên cứu, hoặc soạn thảo bài viết. Hướng dẫn qua nghiên cứu, động não, lập dàn ý, và soạn thảo lặp lại với kiểm soát phiên bản.
日本語の概要は準備中です。原文の説明を表示しています。
根据想法和资料撰写博客草稿。适用于用户想写博客、基于研究创建内容，或起草文章的场景。流程会引导你完成调研、头脑风暴、提纲编写和带版本控制的迭代撰写。
日本語の概要は準備中です。原文の説明を表示しています。
Створення чернетки блог-посту з ідей та ресурсів. Використовуйте, коли користувачі хочуть написати блог-пост, створити контент з дослідження або підготувати статтю. Проводить через дослідження, мозковий штурм, складання плану та ітеративне написання з контролем версій.
日本語の概要は準備中です。原文の説明を表示しています。
确保所有沟通内容都符合品牌语气和风格指南。适用于撰写营销文案、客户沟通、对外内容，或用户提到品牌语气、tone、写作风格的场景。
日本語の概要は準備中です。原文の説明を表示しています。
Ensure all communication matches brand voice and tone guidelines. Use when creating marketing copy, customer communications, public-facing content, or when users mention brand voice, tone, or writing style.
日本語の概要は準備中です。原文の説明を表示しています。