mcpbeat Sign in

Refactoring Surgeon Agent Skill

Expert code refactoring specialist for improving code quality without changing behavior. Activate on: refactor, code smell, technical debt, legacy code, cleanup, simplify, extract method,

13k tokens
context cost
the whole folder, loaded on every use
7
files
ships runnable scripts
0
copies elsewhere
how many repositories repackaged it
177
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/curiositech/some_claude_skills --skill refactoring-surgeon

The instruction itself

23 sections, as written by the author

Refactoring Surgeon

Expert code refactoring specialist focused on improving code quality without changing behavior.

Quick Start

  • Ensure tests exist - Never refactor without a safety net
  • Identify the smell - Name the specific code smell you're addressing
  • Make small changes - One refactoring at a time, commit frequently
  • Run tests after each change - Behavior must remain identical
  • Don't add features - Refactoring ≠ enhancement
  • Document significant changes - Explain the "why" for future maintainers

Core Capabilities

| Category | Techniques |

|----------|------------|

| Extraction | Extract Method, Extract Class, Extract Interface |

| Movement | Move Method, Move Field, Inline Method |

| Simplification | Replace Conditional with Polymorphism, Decompose Conditional |

| Organization | Introduce Parameter Object, Replace Magic Numbers |

| Legacy Migration | Strangler Fig, Branch by Abstraction, Parallel Change |

Code Smells Reference

Bloaters

┌─────────────────────┐    ┌─────────────────────┐    ┌─────────────────────┐
│    Long Method      │    │    Large Class      │    │   Long Parameter    │
│  > 20 lines?        │    │  > 200 lines?       │    │       List          │
│  → Extract Method   │    │  → Extract Class    │    │  → Parameter Object │
└─────────────────────┘    └─────────────────────┘    └─────────────────────┘

OO Abusers

┌─────────────────────┐    ┌─────────────────────┐    ┌─────────────────────┐
│  Switch Statements  │    │   Refused Bequest   │    │   Parallel          │
│  Type-checking?     │    │  Unused inheritance?│    │   Hierarchies       │
│  → Polymorphism     │    │  → Delegation       │    │  → Move Method      │
└─────────────────────┘    └─────────────────────┘    └─────────────────────┘

Change Preventers

┌─────────────────────┐    ┌─────────────────────┐
│  Divergent Change   │    │  Shotgun Surgery    │
│  One class, many    │    │  One change, many   │
│  reasons to change? │    │  classes affected?  │
│  → Extract Class    │    │  → Move/Inline      │
└─────────────────────┘    └─────────────────────┘

Reference Examples

Complete refactoring examples in ./references/:

| File | Pattern | Use Case |

|------|---------|----------|

| extract-method.ts | Extract Method | Long methods → focused functions |

| replace-conditional-polymorphism.ts | Replace Conditional | switch/if → polymorphic classes |

| introduce-parameter-object.ts | Parameter Object | Long params → structured objects |

| strangler-fig-pattern.ts | Strangler Fig | Legacy code → gradual migration |

Anti-Patterns (10 Critical Mistakes)

1. Big Bang Refactoring

Symptom: Rewriting entire modules in one massive change

Fix: Strangler fig pattern, small incremental changes with tests

2. Refactoring Without Tests

Symptom: Changing structure without test coverage

Fix: Write characterization tests first, add coverage for affected areas

3. Premature Abstraction

Symptom: Creating generic frameworks "for future flexibility"

Fix: Wait for three concrete examples before abstracting (Rule of Three)

4. Renaming Without IDE Support

Symptom: Find-and-replace that misses occurrences

Fix: Use IDE refactoring tools, search for usages first

5. Mixing Refactoring and Features

Symptom: Adding new functionality while restructuring

Fix: Separate commits - refactor first, then add features

6. Ignoring Code Reviews

Symptom: Large refactoring PRs that are hard to review

Fix: Small, focused PRs with clear commit messages

7. Over-Abstracting

Symptom: Three layers of abstraction for a simple operation

Fix: YAGNI - start concrete, abstract when patterns emerge

8. Incomplete Refactoring

Symptom: Starting Extract Method but leaving partial duplication

Fix: Complete the refactoring or revert - no half-measures

9. Refactoring Production During Incidents

Symptom: "I'll just clean this up while I'm here..."

Fix: Never refactor during incidents - fix the bug, create a ticket

10. Not Measuring Improvement

Symptom: Refactoring without knowing if it helped

Fix: Track metrics: complexity, test coverage, build time

Safety Checklist

Before Refactoring:

  • [ ] Code compiles/runs successfully
  • [ ] All tests pass
  • [ ] Test coverage is adequate for area being refactored
  • [ ] Commit current state (can rollback)

During Refactoring:

  • [ ] Make small, incremental changes
  • [ ] Run tests after each change
  • [ ] Keep behavior identical
  • [ ] Don't add features while refactoring

After Refactoring:

  • [ ] All tests still pass
  • [ ] No new warnings/errors
  • [ ] Code is more readable
  • [ ] Complexity metrics improved
  • [ ] Document significant changes

Quality Checklist

  • [ ] No behavior changes (tests prove this)
  • [ ] Improved readability
  • [ ] Reduced complexity (cyclomatic, cognitive)
  • [ ] Better adherence to SOLID principles
  • [ ] Removed duplication (DRY)
  • [ ] More testable code
  • [ ] Clear naming
  • [ ] Appropriate abstractions (not over-engineered)

Validation Script

Run ./scripts/validate-refactoring.sh to check:

  • Test coverage presence
  • Code smell indicators
  • Duplication patterns
  • Complexity metrics
  • SOLID violations
  • Refactoring safety (git, uncommitted changes)

External Resources

Other skills for the same job

different authors, same section of the catalogue
Receiving Code Review
by ZhanlinCui
×7

Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation

2k tokens
Requesting Code Review
by ZhanlinCui
×6

Use when completing tasks, implementing major features, or before merging to verify work meets requirements

2k tokens
Git Commit
by github
vendor ×3

Execute git commit with conventional commit message analysis, intelligent staging, and message generation. Use when user asks to commit changes, create a git commit, or mentions "/commit". Supports: (1) Auto-detecting type and scope from changes, (2) Generating conventional commit messages from diff, (3) Interactive commit with optional type/scope/description overrides, (4) Intelligent file staging for logical grouping

799 tokens
Github Code Review
by ComeOnOliver
×3

Comprehensive GitHub code review with AI-powered swarm coordination

13k tokens
Karpathy Guidelines
by hyyhf
×3

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

629 tokens
Code Reviewer
by google-gemini
vendor ×2

Use this skill to review code. It supports both local changes (staged or working tree) and remote Pull Requests (by ID or URL). It focuses on correctness, maintainability, and adherence to project standards.

795 tokens
Agent MD Refactor
by softaworks
×2

Refactor bloated AGENTS.md, CLAUDE.md, or similar agent instruction files to follow progressive disclosure principles. Splits monolithic files into organized, linked documentation.

4k tokens
Commit Work
by softaworks
×2

Create high-quality git commits: review/stage intended changes, split into logical commits, and write clear commit messages (including Conventional Commits). Use when the user asks to commit, craft a commit message, stage changes, or split work into multiple commits.

2k tokens

How to use it

Copy the folder

Take curiositech/refactoring-surgeon from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.