What is the difference between PreToolUse and PostToolUse?

PreToolUse runs before a tool executes to validate or modify inputs, while PostToolUse runs after execution to react to results, provide feedback, or log data.

Can I use hooks to block Claude from finishing a task?

Yes, the 'Stop' hook can evaluate the agent's work and return a 'block' decision with a reason, forcing the agent to continue until the validation requirements are met.

Can I use bash scripts for Claude Code hooks?

Yes, command hooks allow you to execute bash scripts for fast, deterministic validations, file system operations, and integration with external CLI tools.

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

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

What is a prompt-based hook in Claude Code?

A prompt-based hook uses LLM reasoning to evaluate context and make decisions—such as approving or denying a tool call—using natural language instead of traditional code.

Claude Code Hook Development

Name: Claude Code Hook Development
Author: OshanKHZ

byOshanKHZ

•

生产力与工作流

Streamlines the creation and management of event-driven automation, security policies, and workflow validations for Claude Code plugins.

关于

This skill provides a comprehensive framework for implementing lifecycle hooks within Claude Code plugins, allowing developers to automate workflows, enforce security standards, and modify tool behavior in real-time. By supporting both prompt-based logic for context-aware decisions and bash commands for deterministic checks, it enables sophisticated event-driven automation across the entire development lifecycle—from session initialization and tool execution to final task completion. It is an essential resource for developers building robust, production-grade AI agents that require strict governance and project-specific context.

主要功能

Standardized configuration patterns for both global settings and plugin-specific hooks
Deterministic command hooks for bash script execution and system integration
Full lifecycle event coverage including PreToolUse, PostToolUse, and SessionStart
Middleware capabilities to intercept, modify, or block tool inputs and permissions
Advanced prompt-based hooks for context-aware LLM validation and reasoning
3 GitHub stars

使用场景

Implementing quality gates that prevent the agent from finishing tasks until tests pass or criteria are met
Automating project context loading and environment variable persistence at session startup
Enforcing security policies by validating file system operations and bash commands before execution

关于

主要功能

Standardized configuration patterns for both global settings and plugin-specific hooks
Deterministic command hooks for bash script execution and system integration
Full lifecycle event coverage including PreToolUse, PostToolUse, and SessionStart
Middleware capabilities to intercept, modify, or block tool inputs and permissions
Advanced prompt-based hooks for context-aware LLM validation and reasoning
3 GitHub stars

使用场景

Implementing quality gates that prevent the agent from finishing tasks until tests pass or criteria are met
Automating project context loading and environment variable persistence at session startup
Enforcing security policies by validating file system operations and bash commands before execution