How do triggers work in this framework?

Triggers are defined in skill-rules.json and can include explicit keywords, intent-based regex patterns, specific file glob patterns, or content markers within the codebase.

What is the 500-line rule in Claude Code skills?

It is an Anthropic best practice recommending that SKILL.md files remain under 500 lines to ensure the model processes information efficiently; additional details should be placed in reference files.

What is the difference between BLOCK and SUGGEST enforcement?

BLOCK physically prevents tool execution until a specific requirement is met (ideal for critical guardrails), while SUGGEST injects context before the prompt to guide the AI without halting the workflow.

How do I test a new skill trigger manually?

You can manually test hooks by piping a JSON session object into the hook script using 'npx tsx' to verify if the triggers activate as expected for specific prompts or tool inputs.

Skill Developer Guide

Name: Skill Developer Guide
Author: Activer007

byActiver007

0•

Developer Tools

Facilitates the creation, management, and optimization of specialized Claude Code skills following Anthropic's architectural best practices.

The Skill Developer Guide is a comprehensive framework designed for developers who want to extend Claude Code's capabilities through custom skills. It provides structured guidance on implementing the two-hook architecture, managing activation triggers via JSON configuration, and adhering to the 500-line rule for optimal model performance. Whether you are building strict guardrails to prevent database errors or domain-specific advisors for frontend development, this skill ensures your extensions are performant, maintainable, and seamlessly integrated into the Claude environment.

Key Features

01Standardized templates for skill structure and YAML frontmatter

02Advanced trigger configuration using keywords, intent patterns, and file paths

03Optimization strategies through progressive disclosure and session tracking

04Comprehensive guidance on UserPromptSubmit and PreToolUse hook implementation

05Granular enforcement level management (Block, Suggest, Warn)

060 GitHub stars

Use Cases

01Debugging and refining existing skill activation patterns and triggers

02Bootstrapping new domain-specific guidance skills for internal team standards

03Implementing critical safety guardrails to prevent common codebase errors

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add activer007/ordinary-claude-skills skill-developer

For use in Claude.ai and ChatGPT

Download Skill