What is the difference between quick and deep thoroughness?

Quick mode uses fast regex patterns for common smells. Deep mode integrates external CLI tools like jscpd for copy-paste detection and madge for dependency graph analysis to find deeper architectural issues.

What exactly is 'AI slop' in a codebase?

AI slop refers to the unnecessary artifacts frequently left behind by AI assistants, such as 'console.log' statements, placeholder 'Lorem Ipsum' text, 'TODO' comments for already implemented features, and redundant helper functions.

Is it safe to let Deslop automatically modify my code?

Deslop uses a certainty-based system. It only suggests auto-fixes for 'HIGH' certainty items like debug statements. Medium and Low certainty items are reported for manual review to ensure no functional logic is accidentally removed.

Can I run Deslop on just the files I changed?

Yes, by using the --scope=diff argument, Deslop will use Git to identify and scan only the files that have changed in your current branch.

Deslop Code Cleaner

Name: Deslop Code Cleaner
Author: avifenesh

byavifenesh

•

311

•

開発者ツール

Removes AI-generated code smells, debug statements, and unused infrastructure to maintain high-quality repository hygiene.

Deslop is a specialized utility for Claude Code designed to sanitize codebases by identifying and removing 'AI slop'—unnecessary artifacts like leftover debug logs, placeholder text, and orphaned functions often generated during rapid AI development. By utilizing a multi-layered detection pipeline involving regex patterns, AST-based analysis, and repository mapping, it provides a structured report of code smells categorized by certainty and severity. Whether you need a quick scan of your current diff or a deep dive into the entire repository, Deslop helps maintain high standards of code quality by automating the tedious parts of manual cleanup and refactoring.

主な機能

01Certainty-based classification to safely distinguish between auto-fixes and manual reviews

02Flexible scoping to scan specific paths, git diffs, or entire codebases

03Multi-layered detection pipeline using regex and AST-based analysis

04Automatic removal of debug statements, placeholder text, and empty catch blocks

05Repository mapping to identify orphaned infrastructure and unused exports

06311 GitHub stars

ユースケース

01Automated hygiene gates to prevent debug statements from reaching production environments

02Sanitizing AI-generated code before submitting a pull request to ensure professional standards

03Identifying and pruning dead code or unused exports in large, evolving projects

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add avifenesh/awesome-slash deslop

For use in Claude.ai and ChatGPT