What are Claude Code hooks?

Hooks are automated scripts or shell commands that execute at specific points in the Claude Code lifecycle, such as when a session starts or before a tool is used, providing control over the AI's behavior.

Where are hook configurations stored?

Hooks are stored in settings.json files located either in your user home directory (~/.claude/settings.json) or within your project directory (.claude/settings.json).

Can I use Python to write Claude Code hooks?

Yes, this skill provides templates and guidance for writing hooks in both Bash and Python, as long as the script can read JSON from stdin and return appropriate exit codes.

How do hooks improve security?

Hooks can be used as a security layer to validate file paths, sanitize inputs, and prevent dangerous commands from executing by intercepting tool calls before they are processed.

What is a prompt-based hook?

A prompt-based hook uses a natural language prompt instead of a script to make decisions, allowing Claude to decide whether to continue or block an action based on context-aware reasoning.

Claude Code Hook Manager

Name: Claude Code Hook Manager
Author: thoeltig

bythoeltig

Productivity & Workflow

Manages the full lifecycle of Claude Code hooks, from creation and configuration to security validation and performance analysis.

About

The Managing Hooks skill provides a comprehensive framework for orchestrating automated behaviors within the Claude Code environment. By managing lifecycle hooks—scripts that execute at specific points like tool usage, session starts, or prompt submissions—this skill allows developers to enforce deterministic controls, automate repetitive setup tasks, and implement custom security policies. It supports all nine event types, including tool-based and agent-based triggers, ensuring that your AI development workflow is both highly productive and strictly governed by best practices.

Key Features

0 GitHub stars
Automated security validation to prevent shell injection and path traversal vulnerabilities
Interactive workflow for creating, debugging, and updating hook configurations
Support for both command-based (Bash/Python) and prompt-based LLM hooks
Full support for all 9 Claude Code hook event types including PreToolUse and SessionStart
Comprehensive analysis of existing hooks for performance and code quality improvements

Use Cases

Implementing custom safety checks and validation logic before Claude executes file system tools
Auditing and refactoring legacy hook configurations to align with security best practices
Automating environment initialization and cleanup for specific development projects

About

Key Features

0 GitHub stars
Automated security validation to prevent shell injection and path traversal vulnerabilities
Interactive workflow for creating, debugging, and updating hook configurations
Support for both command-based (Bash/Python) and prompt-based LLM hooks
Full support for all 9 Claude Code hook event types including PreToolUse and SessionStart
Comprehensive analysis of existing hooks for performance and code quality improvements

Use Cases

Implementing custom safety checks and validation logic before Claude executes file system tools
Auditing and refactoring legacy hook configurations to align with security best practices
Automating environment initialization and cleanup for specific development projects