What is a Claude Code Skill?

A specialized capability that enhances Claude's behavior for specific tasks by providing domain-specific guidance, best practices, and implementation patterns.

What is the difference between BLOCK and SUGGEST enforcement?

BLOCK prevents tool execution (like file edits) until a condition is met, while SUGGEST injects helpful context into the prompt without stopping the workflow.

How can I bypass a skill guardrail temporarily?

You can use environment variables like SKIP_SKILL_GUARDRAILS, add a // @skip-validation marker to a file, or rely on built-in session tracking that limits repetitive alerts.

How do skill triggers work?

Skills activate based on triggers defined in skill-rules.json, which include explicit keywords, intent-based regex patterns, specific file paths, or file content detection.

What is the 500-line rule in Claude Code?

It is a performance best practice requiring SKILL.md files to stay under 500 lines. Detailed documentation should be moved to reference files to maintain AI efficiency.

Skill Developer Guide

Name: Skill Developer Guide
Author: claudiodearaujo

byclaudiodearaujo

•

开发者工具

Streamlines the creation and management of specialized Claude Code skills using Anthropic's architectural best practices and trigger patterns.

The Skill Developer Guide is a comprehensive framework designed to help developers build, debug, and maintain custom capabilities within Claude Code. It provides a standardized approach to implementing skill triggers—such as keywords, intent patterns, and file paths—while managing complex hook architectures like UserPromptSubmit and PreToolUse. By enforcing industry-standard patterns like the 500-line rule and progressive disclosure, this skill ensures that custom AI guardrails and domain-specific guides remain performant, context-aware, and non-intrusive during the development lifecycle.

主要功能

01Configurable enforcement levels including Block, Suggest, and Warn

021 GitHub stars

03Built-in validation for 500-line rule and progressive disclosure

04Automated skill activation and trigger management

05Dual-hook architecture for proactive suggestions and safety guardrails

06Session-aware state tracking to prevent repetitive notifications

使用场景

01Standardizing domain-specific architectural patterns across a team

02Creating custom guardrails to prevent database schema errors

03Debugging and refining AI activation triggers for complex workflows

What are Skills?·How to Install

Install with 🐟 Skill.Fish

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

For use in Claude.ai and ChatGPT

Download Skill