What happens if I manually edit the generated rule files?

If you use the --update flag, the skill will append new findings while preserving your manual edits. Only the --force flag will overwrite existing files entirely.

How does the skill identify project-specific rules?

The skill uses a combination of file extension counting, configuration file analysis (like package.json or pyproject.toml), and AI-driven pattern recognition to identify unique coding signatures and architectural choices.

Will it expose my private environment variables?

No. The skill uses 'git ls-files' to ensure it only analyzes tracked files, naturally respecting your .gitignore patterns and excluding sensitive untracked files.

Can I customize which directories are analyzed?

Yes, you can configure target directories, exclusions, and language preferences in a .claude/extract-rules.local.md file.

Does it support multiple programming languages?

Yes, it is designed to recognize and generate specific rule sets for TypeScript, JavaScript, Python, Go, Rust, Ruby, Java, and their associated frameworks.

Project Rule Extractor

Name: Project Rule Extractor
Author: hiroro-work

byhiroro-work

•

학습 및 문서화

Extracts project-specific coding standards and domain knowledge from existing codebases to provide structured context for AI agents.

The Project Rule Extractor is a powerful utility designed to bridge the context gap between a codebase and an AI agent. It automatically analyzes source files, project configurations, and documentation to distill architectural patterns, naming conventions, and domain-specific logic into structured markdown files. By generating consistent rule sets for languages and frameworks, it ensures that Claude and other AI tools adhere to your specific implementation patterns, reducing errors and maintaining high code quality without manual documentation maintenance.

주요 기능

01Git-aware scanning that respects .gitignore to protect sensitive data

02Rule extraction from conversation history to capture real-time developer preferences

03Automatic detection of project languages, frameworks, and core dependencies

04Incremental update mode to refresh rules without overwriting manual edits

05Categorization of rules into general style principles and project-specific signatures

0623 GitHub stars

사용 사례

01Standardizing implementation styles across large teams using automated rule generation

02Bootstrapping AI context for a legacy codebase with undocumented patterns

03Capturing and formalizing ephemeral coding rules discussed during an AI chat session

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add hiroro-work/claude-plugins extract-rules

For use in Claude.ai and ChatGPT

Download Skill