Streamlines the creation, management, and optimization of Claude Code skills using standardized architectures, hooks, and guardrail patterns.
概要
This skill provides a comprehensive framework for building and maintaining specialized Claude Code capabilities while strictly adhering to Anthropic’s best practices. It facilitates the implementation of a dual-hook architecture—leveraging UserPromptSubmit for proactive suggestions and Stop hooks for post-response quality checks—ensuring that domain-specific guidance is delivered effectively without disrupting the developer's workflow. By centralizing skill-rules.json management and enforcing the 500-line rule for documentation, it helps developers create robust, session-aware skills that range from critical safety guardrails to deep architectural documentation.
主な機能
Standardized SKILL.md templates optimized for performance and progressive disclosure
Dual-hook architecture for proactive skill suggestions and retrospective error-handling reminders
Automated management of skill-rules.json for precise keyword and intent-based triggers
0 GitHub stars
Comprehensive testing suite for validating skill activation triggers and enforcement levels
Implementation of session-aware guardrails to prevent repetitive or intrusive notifications
ユースケース
Debugging complex skill activation issues and refining regex-based intent patterns
Bootstrapping new domain-specific guidelines with standardized skill templates
Implementing critical safety guardrails to prevent common database or security errors
