ai-slop-cleaner

To clean up AI-generated code slop safely, you must lock existing behavior with regression tests before refactoring. This Skill enforces a tests-first workflow that identifies non-negotiable behavior and strengthens coverage before applying any code changes.

Fallback masking occurs when silent defaults, swallowed errors, or bypass branches hide true failures during code refactoring. This Skill detects these patterns, classifies them, and routes fixes to root-cause remediation or explicit escalation via consensus.

To refactor dead code and duplication in a constrained file scope, use a smell-scoped cleanup plan that applies small, reversible steps. This Skill restricts changes strictly to a provided list of changed files rather than sweeping across whole feature areas.

Yes, you can use this tests-first cleanup workflow for Ralph workflow changes. The Skill specifically supports scope control via file lists, allowing you to apply smell-based refactoring and fallback detection strictly to the changed files within that workflow.

Required validation gates for a minimal diff code refactor include iterative tests, lint, and typecheck. This Skill enforces these quality gates throughout the staged cleanup plan to ensure minimal diffs and generate an information-dense completion report.

You should avoid using an automated smell-based plan for code cleanup when regression test coverage is missing and cannot be established first. This Skill requires non-negotiable behavior to be locked with tests before attempting any deslop or refactoring actions.