refactoring

Refactoring improves code structure through small, isolated changes verified by tests after each step. Apply one change at a time, run your test suite, commit with a clear message, then move to the next change. This cycle ensures behavior stays intact while you eliminate duplication, shorten functions, reduce nesting, and remove code smells.

Refactoring restructures code to improve internal design without changing external behavior; it includes extracting methods, renaming variables, and reorganizing logic. Code cleanup addresses style and formatting. This Skill focuses on refactoring—structural improvements detected through code smells—not formatting or linting.

Refactor when code works but has structural issues: duplicated logic, overly long functions, excessive nesting, or unclear naming. Refactoring preserves proven behavior through mandatory tests after each change, making it safer than rewrites. Rewriting is riskier and better reserved for code too broken to test incrementally.

No. Refactoring should isolate structural improvements from new functionality. Mixing them makes it harder to identify which changes caused test failures and complicates commit history. Separate feature work into its own commits, then refactor in dedicated passes using the one-change-test-commit cycle.

Common code smells include duplicated code blocks, functions longer than one screen, deeply nested conditionals, cryptic variable names, and classes doing too much. This Skill identifies and prioritizes smells, then guides you through extracting methods, renaming, and reorganizing one piece at a time with tests confirming each step.