Claude Code Hook Authoring

关于

This skill serves as a master guide for developers looking to extend Claude's capabilities through event hooks. It covers both declarative JSON configurations for Claude Code and programmatic Python implementations for the Agent SDK, enabling advanced features like tool input validation, automated logging, and context injection. By implementing the provided patterns for security and performance, developers can build robust extensions that enforce project-specific workflows, protect sensitive environments, and maintain audit trails across the agent lifecycle.

主要功能

• Security-first patterns for input validation and secret redaction
• Testing framework for validating hook logic and reliability
• Performance optimization strategies for non-blocking execution
• 10 GitHub stars
• Lifecycle event interception including PreToolUse and PostToolUse
• Declarative JSON vs. programmatic Python SDK implementation guides

使用场景

• Injecting project-specific context into user prompts automatically
• Automating audit logging for compliance and debugging
• Preventing dangerous shell commands in production environments