
Code Refactoring Assistant
Guides safe, test-backed, step-by-step refactoring of existing code using Martin Fowler's methods
- Repository
- luongnv89/claude-howto
- Stars
- 46,401
- Licence
- MIT
- Files
- 4 files
What it does
When you ask the agent to clean up, restructure, or reduce technical debt in code, this skill makes it follow a staged process: first understand the code and check for tests, then list code smells (such as long functions or duplicated logic), propose a written plan of named refactorings with rollback steps, and make small changes one at a time. It asks you to approve each stage and runs the tests after every change so behaviour stays the same.
Great for
- Developers who have legacy code with long functions, duplication or unclear structure and want a careful, reviewable cleanup
- Teams that want refactoring done in small, committed steps with user checkpoints rather than one large rewrite
Skip it if
- Code with no tests and no appetite for risk: the skill itself says not to refactor without tests unless you explicitly accept the risk
- Urgent production incidents: the skill advises against refactoring during an outage
Try asking
- “Refactor the processOrder function in src/orders.js, it is about 150 lines and I want it split up safely”
- “Find code smells in this Python module and propose a prioritized refactoring plan before changing anything”
- “Remove dead code and rename unclear variables in the billing folder, one commit per change”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 4 original files, the MIT licence and an attribution note.
refactor.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip refactor.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip refactor.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Ask the agent to confirm test coverage first, and give it the test command for your project so it can run tests after each step
- Review the proposed plan and approve one stage (quick wins, then structural, then architectural) at a time rather than the whole plan
- Check each commit yourself; the agent's behaviour-preserved claim depends on your tests and manual checks
What's inside
- references/code-smells.md
- references/refactoring-catalog.md
- SKILL.md
- templates/refactoring-plan.md
Redistributed unmodified from luongnv89/claude-howto under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.