Clean AI-generated code slop with regression-safe, deletion-first passes.
npx skills add https://github.com/Rheinmir/skills-kit --skill ai-slop-cleaner-rheinmir
Please help me install this Agent Skill. Skill: ai-slop-cleaner Source: https://github.com/Rheinmir/skills-kit/tree/main/skills/ai-slop-cleaner Command: npx skills add https://github.com/Rheinmir/skills-kit --skill ai-slop-cleaner-rheinmir
Clean AI-generated code slop without drifting scope or changing intended behavior, using a regression-safe, deletion-first workflow.
Provide a bounded anti-slop cleanup on the target files and start with a regression-test pass.
High-intent search queries and answers about installing and using this skill.
AI code slop cleanup preserves intended behavior by enforcing a bounded, deletion-first workflow. It requires a cleanup plan, reuses existing utilities, and verifies all changes with focused regression tests before application to prevent scope drift.
The best way to remove duplicated logic and over-abstracted wrappers is running bounded cleanup passes on working-but-bloated code. This targets weak tests and dead code while keeping diffs small and reversible for safe refactoring.
Yes, you can run a reviewer-only anti-slop pass by using specific flags. This mode generates a clear cleanup plan aligned with existing utilities and patterns without immediately applying changes, ensuring modifications remain reversible and controlled.
Yes, the anti-slop cleanup workflow requires existing regression tests. It starts with a regression-test pass to verify behavior, ensuring that any deletion of dead code or wrapper layers is validated by focused tests before application.
You should avoid using a bounded cleanup pass when you explicitly intend to change application behavior or need large scope expansions. This workflow is designed strictly for working-but-bloated code, prioritizing small, reversible diffs and deletion over addition.
oneforceRemove redundant wrappers and duplicated logic from AI-generated code.
Sir-chawakornRefactor source code in git diffs while preserving functional behavior.
Chang-Shih-YungAggregate type checker, linter, and test results into a 0-10 composite score.
ImL1sRemove AI-generated code slop from Flutter/Dart codebases with regression verification.
