# Refactor

> Improve existing code without changing behavior — split large functions, backfill tests, simplify.

Improve existing code without changing behavior — split large functions, backfill tests, simplify. Use when asked to "refactor", "make robust", "bulletproof", "strengthen", "add missing tests", "split functions", "simplify", "clean up", or "reduce complexity". Don't use for new features (/tdd), security (/harden), or PR review.

- Page: https://ss33.ai/skills/helderberto/refactor-64208515341
- Publisher: helderberto
- Category: [Software Development](https://ss33.ai/skills/category/software-development-t44z8z) › [Testing & QA](https://ss33.ai/skills/category/software-development-t44z8z/testing-and-qa-g56s6z)
- License: MIT
- Source: https://github.com/helderberto/agent-skills/blob/445681fd215e649201f91d65492025f0803efc0a/skills/refactor/SKILL.md

## Skill

SKILL.md: https://ss33.ai/api/v1/ss33/download/helderberto/refactor-64208515341

# Refactor

Target: $ARGUMENTS (file, directory, or module — if blank, use unstaged changes)

Behavior is frozen — the test suite is the contract. Green before and after; if a test's meaning has to change, you've gone too far.

## Workflow

### 1. Scope

- Identify target files (blank → unstaged changes)
- Locate each file's tests using the project's convention (co-located, `__tests__/`, `tests/`, `spec/`). A file with no tests gets **Test gaps** items first — Split/Simplify on untested code is unverifiable, so backfill or skip. No test runner at all → ask which to use. UI flows → backfill end-to-end tests: call the Skill tool with `tdd`, which holds the UI-tests reference. Entrypoint scripts without exports → audit only.
- Run the suite. If red, stop — don't refactor broken code

### 2. Audit

List findings, most-impactful first:

| Bucket | What to look for | Fix |
|---|---|---|
| **Split** | Functions > 20 lines, multiple responsibilities, I/O mixed with logic | Extract pure logic into named helpers; keep I/O at the edges; preserve the original signature |
| **Edge cases** | Missing null/empty/boundary checks at system boundaries, unhandled error paths | Test first, then guard |
| **Test gaps** | Untested public functions, uncovered branches, missing sad-path tests | Backfill through the public interface |
| **Simplify** | Nesting > 2 levels, dead code / unused params, duplicated logic, dense one-liners, shallow pass-through wrappers, speculative config never used, unclear names / primitive obsession | Guard clauses, delete, extract one helper, expand, inline (call the Skill tool with `codebase-design` for the deletion test), remove, rename to intent / small type |

**Chesterton's Fence**: before removing anything, explain why it's there. If you can't, leave it and flag it.

Present the audit as a checklist and ask which items to address — "All items" first, marked (Recommended).

### 3. Apply — one item at a time

Per approved item: RED (failing test exposing the gap, when the item adds behavior coverage) → GREEN (minimal change) → run tests. Real values over mocks; mock only external I/O (call the Skill tool with `tdd` for its mocking guide). If red after the change, revert it and flag as blocked. Never batch.

### 4. Report

Items addressed, tests added, functions extracted, items skipped with reason (Chesterton), suite result before/after.