code-degunker

Detect and fix AI-generated code anti-patterns in files or branch diffs.

45|50|Updated Jan 24, 2026

Search Tags:#code quality #code review #debugging #security #refactoring #ai code

Ready to run

npx skills add https://github.com/zocomputer/skills --skill code-degunker

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: code-degunker
Source: https://github.com/zocomputer/skills/tree/main/Community/code-degunker
Command: npx skills add https://github.com/zocomputer/skills --skill code-degunker

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill identifies and rectifies common anti-patterns found in AI-generated code, ensuring code is production-ready, maintainable, and correct.

Core Features & Use Cases

Anti-Pattern Detection: Scans for 27 distinct AI code anti-patterns across architecture, logic, security, performance, completeness, style, and reliability.
Automated Fixing: Rewrites problematic code sections to eliminate identified anti-patterns while preserving original intent.
Use Case: When you receive code that feels "off" or suspect it was AI-generated, use this Skill to audit it for subtle flaws and automatically refactor it into robust, production-quality code.

Quick Start

Use the code-degunker skill to review and clean up the provided code snippet.

Frequently Asked Questions about code-degunker

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

FAQPage Schema

How do I fix AI-generated code anti-patterns and hallucinated APIs?▼

To fix AI-generated code anti-patterns, you can scan the code to detect hallucinated APIs, incomplete implementations, and missing error handling, then automatically rewrite those sections to restore robustness and production readiness.

What are the most common AI code anti-patterns to look for during a code review?▼

Common AI code anti-patterns include over-engineering, code bloat, dead code, type workarounds, magic numbers, inconsistent patterns, and poor architecture, alongside subtle security vulnerabilities and performance issues.

Can I review a full branch diff for security vulnerabilities and missing error handling?▼

Yes, you can review a full branch diff or a single file to detect missing error handling and security vulnerabilities, ensuring the AI-generated code logic is correct and maintainable.

How do I automatically refactor code bloat and dead code from AI-generated scripts?▼

You can automatically refactor code bloat and dead code by identifying these anti-patterns and rewriting the problematic sections, which eliminates unnecessary complexity while preserving the original intent.

Does the code degunker work on incomplete implementations and TODOs left by AI?▼

Yes, code-degunker identifies incomplete implementations and TODOs left by AI, rectifying these flaws to ensure the final code is correct, maintainable, and fully production-ready.

What's the best way to audit suspect AI code for subtle flaws before production deployment?▼

The best way to audit suspect AI code is to scan for 27 distinct anti-patterns across architecture, logic, security, performance, completeness, style, and reliability, automatically refactoring the flaws into robust code.