Hooks run with your local user permissions. For security, it is best practice to use absolute paths for scripts and sanitize inputs before execution.

What are Claude Code hooks?

Hooks are shell commands that execute automatically at specific points in Claude Code's lifecycle to provide deterministic control over the AI's behavior.

Can I use languages other than Bash for hooks?

Yes, you can use any language that can be executed from the shell, such as Python or Node.js, as long as it can parse JSON from stdin.

Where are hook configurations stored?

Hooks are configured in the settings.json file, located either globally in ~/.claude/ or locally within a project's .claude/ directory.

How do I block an action using a hook?

You can block a tool operation by having your hook exit with code 2 or by returning a JSON object with the decision field set to 'block'.

Claude Code Hooks Creator

Name: Claude Code Hooks Creator
Author: hongbietcode

byhongbietcode

•

Productivity & Workflow

Automates and manages Claude Code lifecycle events by creating custom shell hooks for task automation, file protection, and code formatting.

This skill empowers developers to extend Claude Code's capabilities through custom lifecycle hooks, enabling deterministic control over the AI's behavior. It provides a comprehensive framework for implementing triggers at critical stages—such as before or after tool use, and during session starts or ends—to enforce file protection rules, automate code formatting, and maintain session state. By moving repetitive tasks from LLM prompting to automated shell commands, users can ensure consistency, security, and efficiency in their AI-assisted development workflows.

Key Features

01Customizable JSON-based input/output for advanced logic control

02Lifecycle trigger management for PreToolUse, PostToolUse, and Stop events

03Seamless integration with code formatters like Prettier and Black

042 GitHub stars

05Automated file protection and sensitive path edit blocking

06Production-ready templates for memory reminders and state management

Use Cases

01Preventing the AI from editing sensitive configuration files or credentials

02Automatically running linters or formatters after the AI modifies source code

03Enforcing persistent memory by prompting the AI to store learnings at session end

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add hongbietcode/synthetic-claude cc-hooks-creator

For use in Claude.ai and ChatGPT

Download Skill