[AUTO-REFACTOR REPORT]
Dedicated documentation for the /refactor command.
// Scan the project, identify large/complex files, and produce a refactor plan.
$ /refactor
This command analyzes your repository to find oversized or complex files, components, or pages. It then generates a concise, actionable plan to split them into smaller, reusable parts while preserving behavior.
How it works
- Scans source directories (e.g.,
src/
) and measures file size and basic complexity signals. - Flags candidates like long React components, pages with many responsibilities, or utility files doing too much.
- Outputs a prioritized list with suggested extractions and proposed component/function names.
What is considered a 'big file'
- High line count or many nested blocks.
- Multiple responsibilities (rendering + data fetching + state + side effects).
- Duplicated patterns that could be abstracted into shared components/hooks.
Report format
- File: path and quick metrics.
- Issues: why it was flagged (size, complexity, cohesion).
- Proposed plan: step-by-step extractions into smaller components, hooks, or modules.
- Reusability targets: where new pieces can be shared (e.g.,
src/components/ui
).
Usage
- Open QLOOD-CLI and run
/refactor
. - Review the generated AutoRefactor report and follow the suggested steps.
- Iterate: extract small pieces, re-run to validate improvements.
See also: [COMMANDS] · [AUTO-REVIEW]