skills.rest
Uniquecrete
UniquecreteCOMMUNITY
Explore all skills by this author →
GitHub

code-degunker

Review and refactor AI-generated code to fix anti-patterns and security issues.

Updated May 6, 2026
Search Tags:#code review#refactoring#error handling#production readiness#security hardening#anti-pattern detection#hallucinated APIs
One-click install
npx skills add https://github.com/Uniquecrete/ThinkFasterv1 --skill code-degunker-uniquecrete
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-degunker
Source: https://github.com/Uniquecrete/ThinkFasterv1/tree/main/Skills/code-degunker
Command: npx skills add https://github.com/Uniquecrete/ThinkFasterv1 --skill code-degunker-uniquecrete

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill identifies and fixes AI-generated code anti-patterns that look plausible but fail under real-world conditions, including incomplete implementations, hallucinated APIs, missing error handling, security vulnerabilities, performance issues, and maintainability problems.

Core Features & Use Cases

  • Anti-pattern detection and triage: Scans for 27 categories of issues across architecture, logic, security, performance, completeness, style, and reliability, then ranks findings by severity from Critical to Minor.
  • Rewrite instead of only flagging: Refactors problematic sections to preserve intent while removing slop and simplifying the solution.
  • Production-shipping review mode: Includes a final “would I ship this?” pass to ensure the output is maintainable, correct, and safe to deploy.
  • Use cases: Use it to degunk code you suspect is AI-generated, clean up “working but wrong” implementations, harden insecure handlers and database queries, and make code reviews faster by converting review comments into concrete fixes.

Quick Start

Use the code-degunker skill to review the provided file for AI-generated slop and return a patched, production-ready version with the critical issues fixed first.

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 that contains hallucinated APIs and insecure logic?▼

You can harden AI-generated code by scanning for 27 anti-pattern categories, including hallucinated dependencies and insecure logic. The process triages findings by severity, rewrites problematic sections while preserving intent, and applies a final production-shipping review pass.

What is the best way to review a full file for AI code anti-patterns and missing error handling?▼

The best way is a full-file review that identifies weak error handling and incomplete implementations across architecture, logic, and reliability. It flags issues by severity from Critical to Minor and rewrites fragile structures into maintainable, production-ready code.

Can I use this refactoring approach on a diff or branch review scenario?▼

Yes, this approach applies to both full-file and diff/branch review scenarios. It evaluates code that feels production-unready or over-engineered, converting review findings into concrete fixes that validate correctness and security readiness for shipping.

How does anti-pattern detection handle security hardening and performance issues?▼

Anti-pattern detection scans for insecure database queries and inefficient structures, ranking them by severity from Critical to Minor. It then refactors the problematic code to reduce complexity, remove AI slop, and ensure the output is safe to deploy.

When should I not use automated refactoring for production readiness?▼

You should avoid automated refactoring when the original implementation intent is ambiguous or requires complex domain-specific validation. It is designed for removing AI slop and hardening fragile structures, not for architecting entirely new systems from scratch.

Related Skills

Explore More Software Engineering Skills→
zocomputerzocomputer
45

code-degunker

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

Official
Advanced
heaptracetechnologyheaptracetechnology
13

code-review

Review code changes for logic errors, security issues, and anti-patterns.

Community
Advanced
LuisSambranoLuisSambrano
7

code-review

Perform systematic code review and static analysis across multiple programming languages.

Community
Advanced
AvihusittonAvihusitton

logic-lens

Detect logic flaws, race conditions, and security risks in code reviews.

Community
Intermediate