Skip to content

About

Three domain-specific code quality assurance tools for Claude Code — generic, AI/ML, and quantitative trading. Based on Clean Code heuristics.

Topics

Resources

Stars

2 stars

Watchers

0 watching

Forks

Latest commit

 

History

7 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

clean-code-audit

Three domain-specific code quality assurance skills based on Robert C. Martin's Clean Code heuristics. Each version extends generic Clean Code checks with domain-specific bad-smell detection.

Structure

clean-code-audit/
├── README.md
├── LICENSE
├── generic/
│   ├── SKILL.md
│   └── scanner.py
├── ai/
│   ├── SKILL.md
│   └── scanner.py
└── quant/
    ├── SKILL.md
    └── scanner.py

Clean Code Digest

The full Clean Code book has been distilled into a bilingual Chinese-English theme-based digest in the docs/ directory — 15 theme files covering all 17 chapters + 3 appendices.

Use it alongside the scanners as a reference: when the scanner flags an issue with a heuristic ID (e.g. G30), look it up in the digest for the full context.

Version Scanner Domain-Specific Heuristics
Generic generic/scanner.py Clean Code heuristics only (G1–G36, N1–N7, F1–F4, etc.)
AI/ML ai/scanner.py Data leakage, missing seeds, hardcoded hyperparameters, model versioning (AI1–AI8)
Quant quant/scanner.py Lookahead bias, hardcoded tickers, transaction costs, signal-position coupling (Q1–Q8)

Quick Start

Generic

python3 generic/scanner.py --file src/main.py
python3 generic/scanner.py --dir ./project --ext .py,.java
python3 generic/scanner.py --git-diff

AI/ML

python3 ai/scanner.py --file train.py
python3 ai/scanner.py --dir ./ml-project --ext .py
python3 ai/scanner.py --git-diff

Quant

python3 quant/scanner.py --file backtest.py
python3 quant/scanner.py --dir ./quant-project --ext .py
python3 quant/scanner.py --git-diff

Installing via npx

From any machine with GitHub access, install all three skills in one command:

npx skills add elonwoo-02/clean-code-audit --global --yes

Or install a single version:

npx skills add elonwoo-02/clean-code-audit@code-quality --global --yes
npx skills add elonwoo-02/clean-code-audit@code-quality-ai --global --yes
npx skills add elonwoo-02/clean-code-audit@code-quality-quant --global --yes

This installs the skill symlinks into ~/.agents/skills/ and makes them available to Claude Code, Cursor, Codex, GitHub Copilot, and other compatible agents.

If you cannot reach GitHub from your machine (network/firewall), clone the repo first and install from the local path:

git clone https://github.com/elonwoo-02/clean-code-audit.git
npx skills add --global --yes ./clean-code-audit

Installing manually (copy method)

Each subdirectory is a standalone skill — copy its contents into a ~/.claude/skills/<name>/ folder.

# Generic
mkdir -p ~/.claude/skills/code-quality
cp generic/SKILL.md ~/.claude/skills/code-quality/
cp generic/scanner.py ~/.claude/skills/code-quality/

# AI/ML
mkdir -p ~/.claude/skills/code-quality-ai
cp ai/SKILL.md ~/.claude/skills/code-quality-ai/
cp ai/scanner.py ~/.claude/skills/code-quality-ai/

# Quant
mkdir -p ~/.claude/skills/code-quality-quant
cp quant/SKILL.md ~/.claude/skills/code-quality-quant/
cp quant/scanner.py ~/.claude/skills/code-quality-quant/

Zero Dependencies

All three scanners use only Python 3 standard library (ast, re, argparse). No pip install needed.

License

MIT

About

Three domain-specific code quality assurance tools for Claude Code — generic, AI/ML, and quantitative trading. Based on Clean Code heuristics.

Topics

Resources

Stars

2 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages