skills.rest
dtbuchholz
dtbuchholzCOMMUNITY
Explore all skills by this author →
Website𝕏@dtbuchholzGitHub

deslop

Analyze diffs against the main branch to remove AI-generated code slop.

Updated Jan 22, 2026
Search Tags:#code quality#developer tools#code review#refactoring#cleanup#ai code
One-click install
npx skills add https://github.com/dtbuchholz/claude-config --skill deslop-dtbuchholz
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deslop
Source: https://github.com/dtbuchholz/claude-config/tree/main/skills/deslop
Command: npx skills add https://github.com/dtbuchholz/claude-config --skill deslop-dtbuchholz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automatically identifies and removes common patterns of "AI slop" in code, such as unnecessary comments, overly defensive programming, and type hacks, ensuring cleaner, more human-readable code.

Core Features & Use Cases

  • AI Code Cleanup: Detects and removes comments that merely state the obvious.
  • Defensive Code Reduction: Eliminates redundant error handling and unnecessary checks.
  • Type Hack Removal: Identifies and suggests removal of any casts and @ts-ignore comments.
  • Use Case: After an AI assistant generates code for a new feature, use this Skill to refine it before committing, making it indistinguishable from human-written code.

Quick Start

Use the deslop skill to clean up AI-generated code slop in the current branch.

Frequently Asked Questions about deslop

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I clean up AI-generated code slop from recent changes?▼

Cleaning up AI-generated code slop involves analyzing recent diffs against the main branch to remove unnecessary comments, over-defensive code, and type hacks, resulting in human-readable code.

What is AI code slop and how do I identify it in my codebase?▼

AI code slop consists of unnecessary comments stating the obvious, overly defensive programming with redundant error handling, and type hacks like `any` casts or `@ts-ignore` comments that make code inconsistent with human-written standards.

How do I remove unnecessary comments and defensive code after using an AI assistant?▼

You can remove unnecessary comments and defensive code by running a cleanup process that detects obvious statements and redundant error handling, refining the AI-generated output before committing to make it indistinguishable from human-written code.

Does AI code cleanup work on diffs against the main branch?▼

Yes, AI code cleanup works by analyzing diffs against the main branch, launching parallel detection agents for each changed file to efficiently identify and remove style inconsistencies and type hacks.

Can I use this to remove type hacks like any casts and ts-ignore comments?▼

Yes, you can remove type hacks by using this Skill to identify and suggest the removal of `any` casts and `@ts-ignore` comments, eliminating style inconsistencies that differ from human-written code standards.

What are the limitations of using parallel detection agents for code refactoring?▼

The parallel detection agents operate using the 'haiku' model for efficient analysis, focusing specifically on recent changes and identifying patterns like unnecessary comments and type hacks rather than performing comprehensive architectural refactoring.

Related Skills

Explore More Software Engineering Skills→
chrisliu298chrisliu298
1

deslop

Remove AI-generated commentary and dead code from git diffs.

Community
Intermediate
ntk148vntk148v
1

conventional-commits

Generate conventional commit messages with type, scope, and breaking change markers.

Community
Basic
lehuuhau1231lehuuhau1231

lint-and-validate

Run linting and type checking across multiple programming ecosystems.

Community
Intermediate
deadlock-mod-managerdeadlock-mod-manager
408

deslop

Remove unnecessary AI-injected comments and defensive checks from source code files.

Official
Intermediate