Remove duplication, dead code, weak types, and legacy paths from codebases.
npx skills add https://github.com/arjitj2/arjit-skills --skill deslop-arjitj2
Please help me install this Agent Skill. Skill: deslop Source: https://github.com/arjitj2/arjit-skills/tree/main/skills/deslop Command: npx skills add https://github.com/arjitj2/arjit-skills --skill deslop-arjitj2
This Skill helps you systematically improve an existing codebase when it has duplicated logic, weak typing, unused code, circular dependencies, legacy branches, defensive fallbacks, or AI-generated clutter that makes the project harder to maintain.
any-heavy models, and uses this Skill to reduce complexity while preserving behavior.Use the deslop skill to analyze the repository for dead code, weak types, legacy paths, and duplicated logic, then implement only the highest-confidence cleanup changes and verify them with the project’s checks.
High-intent search queries and answers about installing and using this skill.
To remove AI slop and clean up a legacy codebase, you can coordinate an eight-lane analysis workflow covering deduplication, dead code, weak types, circular dependencies, and placeholder content, applying only high-confidence refactoring changes verified by typecheck and lint.
The best way to safely delete dead code is by inspecting the dependency graph and running static analysis to identify unused paths, then verifying removal with targeted tests and build validation to preserve existing behavior.
Dependency graph inspection helps code cleanup by mapping import cycles and unused modules, enabling evidence-based refactoring that eliminates circular dependencies while confirming structural integrity through type validation.
Yes, repository-wide maintenance specifically targets weak typing and any-heavy models by analyzing type quality lanes, strengthening shared types, and verifying improvements through typecheck and lint validation.
You should avoid speculative rewrites during code refactoring when lacking sufficient repository evidence, prioritizing targeted research and static analysis to ensure high-confidence changes that pass typecheck and tests instead.
zcaceresAnalyze software projects for dead code, duplicates, and circular dependencies.
AnveioRun and debug repository verification pipelines with ./verify.sh and E2E modes.
jacexhFlag common Python anti-patterns in source code during reviews and debugging.
winsorllcAnalyze codebases for structural patterns, complexity metrics, and dependency graphs.