Clean Code Refactoring Assistant
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md to install @org-02qudk26/code-refactoring-refactor-clean into your AI assistant.
About this skill
Problem
When code has duplication, deep nesting, mixed responsibilities, unclear dependencies, or weak tests, adding more features can amplify the mess. This is better used for module-level refactoring rather than single-line fixes, documentation cleanups, or temporary changes during a change freeze. For code with heavy cross-module coupling or fuzzy boundaries, it is useful to map risks before deciding how to split responsibilities.
How It Works
The skill follows a diagnose-plan-incremental-apply-verify workflow:
- Assess code smells: identify duplication, high complexity, risk hotspots, dependency inversion, and hard-to-test boundaries.
- Create an incremental plan: define target areas, ordered steps, and expected impact, avoiding large one-shot rewrites.
- Keep behavior stable: apply small reviewable diffs, and use resources/implementation-playbook.md when deeper patterns are needed.
- Verify regressions: update tests and confirm original behavior remains unchanged; external behavior should not change without explicit approval.
- Produce executable output: typically includes a problem summary, target areas, refactoring steps, suggested changes, expected impact, and validation notes.
Boundary: Do not trigger a full refactor if the request is only to explain code, fix a bug, or add documentation; if the project forbids structural changes, confirm the allowed scope first.
Use Cases
- Take over a business module with deep nesting and duplicated logic, and need a reviewable incremental refactoring plan.
- Prepare a module for a new payment flow, and need to reduce circular dependencies while adding test coverage for critical paths.
- Code review finds an overly long function with mixed responsibilities, and it needs to be split according to SOLID with impact notes.
- Before launch, changes are frozen, so the request is limited to a problem summary, risk points, and the smallest safe change.
Best For
- Backend engineers maintaining legacy business code, who want to identify code smells before incrementally splitting service boundaries.
- Tech leads planning refactoring work, who need to turn code issues into reviewable and verifiable incremental plans.
- Platform engineers preparing to add features to older modules, who want to reduce coupling and strengthen tests first.
- Senior engineers doing code review, who expect refactoring advice to keep behavior stable and keep diffs reviewable.
Related Skills
Guides Java Spring Boot backend development across REST APIs, data access, security, microservices, testing, deployment, and troubleshooting.
A systematic code error fixing skill that follows a five-step workflow: Reproduce, Isolate, Understand, Fix, Verify, to help developers resolve code errors.
Restores Lanhu designs to target framework code using schema JSON, design tokens, and image assets.
Automatically detects engineering or competition code, analyzes bugs, vulnerabilities, algorithm complexity, and edge cases, then produces risk-level findings with before/after fixes.