How does this skill help with code reviews?

It provides specialized agents for testing, dead code, abstraction, and style, allowing Claude to automatically flag specific quality issues that humans might overlook during rapid iteration.

What reporting format does the skill use?

The skill reports issues using a standardized [FILE:LINE] ISSUE_TITLE format, making it easy to integrate into automated workflows or manual fix-it lists.

Can I use this for non-AI generated code?

Yes, while optimized for AI-introduced patterns, the detection criteria for dead code, over-abstraction, and test quality are applicable to any professional software project.

Which programming languages is this compatible with?

The logic is language-agnostic but particularly effective for Python, Go, React, and other frameworks where LLM agents are frequently used.

What are 'LLM artifacts' in a codebase?

LLM artifacts are patterns commonly introduced by AI agents, such as overly verbose comments, redundant test setups, and over-engineered abstraction layers that can bloat a codebase.

LLM Artifacts Detection

Name: LLM Artifacts Detection
Author: existential-birds

byexistential-birds

•

보안 및 테스팅

Identifies and cleans up AI-generated code patterns, including redundant tests, dead code, and over-engineered abstractions.

소개

The LLM Artifacts Detection skill is designed to maintain high code quality by identifying 'AI cruft'—the specific types of technical debt often introduced by LLM coding agents. It deploys specialized sub-agents to scan for DRY violations in tests, orphaned code fragments, unnecessary layers of abstraction, and verbose documentation styles that restate the obvious. By providing a structured framework for code verification, this skill ensures that AI-accelerated development doesn't lead to long-term maintainability issues or codebase bloat.

주요 기능

Flagging of verbose 'LLM-style' comments and defensive overkill
Detection of DRY violations and shallow mock boundaries in test suites
14 GitHub stars
Structured reporting with file and line-specific issue summaries
Analysis of over-abstraction patterns like redundant wrapper classes
Identification of unused code, TODOs, and legacy backwards-compatibility cruft

사용 사례

Refactoring legacy codebases to remove 'copy-paste drift' and redundant layers
Post-generation cleanup of code produced by AI coding assistants
Improving test suite quality by identifying shallow mocks and library testing

소개

주요 기능

Flagging of verbose 'LLM-style' comments and defensive overkill
Detection of DRY violations and shallow mock boundaries in test suites
14 GitHub stars
Structured reporting with file and line-specific issue summaries
Analysis of over-abstraction patterns like redundant wrapper classes
Identification of unused code, TODOs, and legacy backwards-compatibility cruft

사용 사례

Refactoring legacy codebases to remove 'copy-paste drift' and redundant layers
Post-generation cleanup of code produced by AI coding assistants
Improving test suite quality by identifying shallow mocks and library testing