hookify

Manage hookify rules for Claude behaviors using bash, file, stop, and prompt events.

12|2|Updated Nov 20, 2025
One-click install
npx skills add https://github.com/alexandrbasis/wythm-claude-workflows --skill hookify-alexandrbasis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hookify
Source: https://github.com/alexandrbasis/wythm-claude-workflows/tree/main/.claude/skills/hookify
Command: npx skills add https://github.com/alexandrbasis/wythm-claude-workflows --skill hookify-alexandrbasis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps prevent unwanted or unsafe behaviors from Claude by allowing you to define and manage custom rules that intercept and modify its actions.

Core Features & Use Cases

  • Rule Creation: Define rules based on events like bash commands, file operations, user prompts, or agent stops.
  • Behavior Control: Block or warn against specific actions, ensuring compliance and safety.
  • Use Case: Prevent Claude from accidentally deleting important files by creating a rule that blocks any rm -rf command targeting sensitive directories.

Quick Start

Use the hookify skill to list all currently configured rules.

Frequently Asked Questions about hookify

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent an AI agent from running unsafe bash commands automatically?

To prevent unsafe bash commands, you can define custom behavior control rules that intercept agent actions and block execution based on specific patterns. This ensures compliance by evaluating commands against configured safety conditions before they run.

What is prompt behavior control and how does it work for AI compliance?

Prompt behavior control works by defining custom rules that intercept and modify AI actions during events like file operations or bash commands. It uses regex pattern matching to evaluate conditions and block or warn against actions that violate compliance rules.

Can I block specific file operations like deleting sensitive directories during automated tasks?

Yes, you can block specific file operations by creating rules that target file events. For example, you can configure a rule to intercept and block any command targeting sensitive directories, preventing accidental deletion of important files during automated tasks.

How do I set up custom rules to warn against unwanted file operations?

You can set up custom rules by defining conditions based on file events and using regex pattern matching to specify unwanted operations. The system evaluates these conditions during file operations and triggers a warning instead of executing the action.

What types of events can I intercept to enforce AI safety rules?

You can intercept events including bash commands, file operations, user prompts, and agent stops. This allows you to evaluate conditions using regex pattern matching across different stages of interaction to enforce comprehensive safety and compliance rules.

How do I analyze past conversations to identify behaviors that need compliance rules?

You can analyze past conversations to identify potential behaviors to hook by reviewing the interaction history. This analysis helps you discover unwanted actions that require new behavior control rules to ensure future compliance and safety.