What is the 500-line rule for Claude skills?

It is a best practice to keep SKILL.md files under 500 lines to ensure Claude can efficiently process the context; deeper details should be moved to reference files.

How do BLOCK and SUGGEST enforcement levels differ?

BLOCK prevents tool execution (like editing a file) until requirements are met, whereas SUGGEST simply adds helpful context to the prompt without stopping the workflow.

Can I disable specific skill guardrails?

Yes, you can use environment variables like SKIP_SKILL_GUARDRAILS or add '// @skip-validation' markers to specific files to bypass skill checks.

What is a Claude Code Skill?

A Claude Code Skill is a specialized capability defined in Markdown and JSON that enhances Claude's behavior, providing it with domain-specific guidance and automated triggers.

Claude Skill Developer

Name: Claude Skill Developer
Author: claudiodearaujo

byclaudiodearaujo

0•

開発者ツール

Orchestrates the creation, configuration, and management of custom Claude Code skills using Anthropic best practices.

The Claude Skill Developer is a comprehensive framework designed for users who want to extend Claude's capabilities with specialized domain knowledge and automated guardrails. It provides a structured approach to building skills using a two-hook architecture—UserPromptSubmit for proactive suggestions and PreToolUse for enforcement—ensuring that Claude follows specific project patterns, coding standards, and error-handling protocols. By implementing the 500-line rule and progressive disclosure patterns, this skill allows developers to create high-performance, context-aware AI assistants that adapt to any technical stack or organizational workflow.

主な機能

01Multi-level enforcement options including BLOCK (guardrails) and SUGGEST (advisory)

02Automated skill activation through keyword, intent, and file-path trigger patterns

03Session-aware state management to prevent repetitive or intrusive reminders

04Dual-hook system for proactive prompt injection and reactive tool enforcement

05Best practice enforcement for skill documentation and progressive disclosure architectures

060 GitHub stars

ユースケース

01Establishing project-wide guardrails to prevent critical database or API integration errors

02Automating the delivery of domain-specific documentation based on the developer's current file context

03Creating custom linting and architectural reviews that trigger before code is committed or executed

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add claudiodearaujo/sistema-de-narra-o-de-livro-front skill-developer

For use in Claude.ai and ChatGPT

Download Skill