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

The 500-line rule is an Anthropic best practice recommending that main SKILL.md files remain under 500 lines to maintain focus and context window efficiency, utilizing reference files for more granular details.

What is the purpose of the UserPromptSubmit hook?

The UserPromptSubmit hook allows the system to proactively suggest relevant skills to Claude before it processes a user's prompt, injecting context based on detected intent.

How can I debug a skill that isn't activating?

You can debug activation by setting the SKILL_DEBUG=true environment variable or using the 'yarn lint:skill' and 'dev:skill' commands to view the underlying matching logic.

How do triggers work in Claude Code skills?

Skills are activated via keyword matching (exact substrings) or intent patterns (regex matching) defined in the skill-rules.json configuration file.

Writing Claude Code Skills

Name: Writing Claude Code Skills
Author: Activer007

byActiver007

学习与文档

Provides comprehensive guidance for creating, managing, and debugging custom skills in Claude Code following Anthropic's official best practices.

关于

This skill serves as a meta-framework for developers looking to extend Claude Code's capabilities. It provides the architectural patterns needed to build efficient, auto-activating skills, covering everything from YAML frontmatter and the 500-line rule to complex regex intent patterns and hook system mechanics. By implementing progressive disclosure and standardized validation techniques, it ensures that custom AI enhancements remain maintainable, performant, and perfectly aligned with Claude's operational constraints.

主要功能

Standardized YAML frontmatter and metadata structure
UserPromptSubmit hook implementation mechanics
Automated skill activation debugging tools
Anthropic 500-line rule enforcement guide
Custom trigger and intent pattern templates
0 GitHub stars

使用场景

Building domain-specific knowledge bases for specialized engineering teams
Standardizing AI behavior and documentation patterns across a codebase
Debugging and refining complex skill activation triggers in Claude Code

关于

主要功能

Standardized YAML frontmatter and metadata structure
UserPromptSubmit hook implementation mechanics
Automated skill activation debugging tools
Anthropic 500-line rule enforcement guide
Custom trigger and intent pattern templates
0 GitHub stars

使用场景

Building domain-specific knowledge bases for specialized engineering teams
Standardizing AI behavior and documentation patterns across a codebase
Debugging and refining complex skill activation triggers in Claude Code