# Refactor

> Safely refactor code while maintaining all tests and type safety (Daniel Okoye's workflow)

- Skill: `majiayu000/refactor-5` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds add majiayu000/refactor-5`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majiayu000/refactor-5/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Integrations & APIs
- Author: majiayu000 (https://skillmd.com/u/majiayu000)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/majiayu000/refactor-5

---


Refactor: $ARGUMENTS

Follow Daniel Okoye's architecture standards:

1. **Baseline**: Run the test and type-check commands (see project config) BEFORE making any changes.
   If either fails, fix existing issues first — never refactor on a broken baseline.

2. **Analyze**: Read the code to be refactored and all its callers
   - Map the dependency graph: who imports/calls this code?
   - Identify the layer: Core → Service → API
   - Check for test coverage of current behavior

3. **Refactor rules**:
   - Preserve the public API — same function signatures, same return types
   - If renaming, update ALL callers and tests in the same commit
   - Maintain layer boundaries (see Layers in project config) — never reverse dependency flow
   - Keep route handlers thin — if moving logic, move it INTO service layer, not out
   - All service access still via DI injection (see stack concepts) — no direct imports in API layer
   - Keep future annotations pattern in every file (see stack concepts)
   - No new untyped/any types — use explicit types

4. **If splitting files**:
   - Update `__init__.py` exports if needed
   - Ensure no circular imports between layers
   - Each new file follows the future annotations pattern (see stack concepts)

5. **If renaming**:
   - Search entire codebase for old name: source files, tests, config, docs
   - Update project documentation if it references renamed items
   - Update OpenAPI metadata (summary, description) if endpoints changed

6. **Verify**: Run the test and type-check commands again.
   All tests MUST still pass. No new type errors. No regressions.