How do I ensure my hooks are portable across different machines?

Always use the ${CLAUDE_PLUGIN_ROOT} environment variable in your hook configurations to reference scripts relative to the plugin directory rather than absolute paths.

Do multiple hooks for the same event run in order?

No, all matching hooks for a specific event run in parallel. They are independent and do not see the outputs of other hooks running at the same time.

What is the difference between prompt-based and command hooks?

Prompt-based hooks use natural language reasoning for context-aware decisions, while command hooks execute bash scripts for fast, deterministic file system operations and tool integrations.

Can I block a tool from executing using a hook?

Yes, by using a PreToolUse hook, you can return a 'deny' decision or exit with code 2 to prevent a tool from running and provide feedback to Claude.

What is the purpose of the SessionStart hook?

SessionStart allows you to execute scripts when a Claude Code session begins, which is ideal for loading project context, setting environment variables, or preparing the workspace.

Claude Code Hook Development

Name: Claude Code Hook Development
Author: nordeim

bynordeim

開発者ツール

Streamlines the creation and implementation of event-driven automation hooks for Claude Code plugins.

概要

This skill provides a comprehensive framework for developers to extend Claude Code through specialized event-driven hooks. It enables the implementation of both deterministic command-based checks and advanced prompt-based validation, allowing developers to intercept tool calls, enforce security policies, and automate complex workflows. Whether you're building guardrails for dangerous commands or setting up project-specific context during session start, this skill offers the implementation patterns and best practices needed to create robust, production-grade Claude Code integrations.

主な機能

Completion verification logic using Stop and SubagentStop hooks
0 GitHub stars
Comprehensive tool validation via PreToolUse and PostToolUse events
Support for context-aware Prompt-Based and deterministic Command-Based hooks
Automated session lifecycle management with SessionStart and SessionEnd
In-depth security patterns for path safety and input sanitization

ユースケース

Automating environment setup and project context loading at the start of every AI session
Implementing custom AI-driven quality gates to verify code completion before the agent stops
Building automated security guardrails to block dangerous shell commands or path traversal

概要

主な機能

Completion verification logic using Stop and SubagentStop hooks
0 GitHub stars
Comprehensive tool validation via PreToolUse and PostToolUse events
Support for context-aware Prompt-Based and deterministic Command-Based hooks
Automated session lifecycle management with SessionStart and SessionEnd
In-depth security patterns for path safety and input sanitization

ユースケース

Automating environment setup and project context loading at the start of every AI session
Implementing custom AI-driven quality gates to verify code completion before the agent stops
Building automated security guardrails to block dangerous shell commands or path traversal