This skill provides comprehensive guidance for developers building Claude Code plugins, focusing on the implementation of both prompt-based and command-based hooks. It enables users to intercept tool calls (PreToolUse), react to results (PostToolUse), enforce completion standards (Stop), and manage session contexts (SessionStart/End). By offering standardized patterns for validation, security best practices, and configuration formats, this skill ensures that plugins are robust, secure, and seamlessly integrated into the Claude Code environment, allowing for complex, context-aware automation workflows.
주요 기능
01Security best practices for path safety and input validation in plugin scripts
02Automated session lifecycle management including SessionStart and SessionEnd
03Step-by-step guidance for PreToolUse and PostToolUse validation interceptors
04Implementation patterns for advanced LLM-driven prompt-based hooks
05Deterministic check configuration using Bash-based command hooks
061 GitHub stars
