What types of events can I trigger rules on?

You can trigger rules on bash commands, file edits (Write/Edit/MultiEdit), stop events (when the agent finishes), and user prompts.

Can I prevent Claude from running dangerous commands?

Yes, by setting the event to 'bash' and the action to 'block' for patterns like 'rm -rf', you can stop the execution of hazardous terminal commands.

What is a Hookify rule?

A Hookify rule is a specialized Markdown file with YAML frontmatter that instructs Claude Code to watch for specific patterns and respond with warnings or blocks when those patterns are matched.

Do I need to restart Claude for rule changes to take effect?

No, rules are read dynamically and updated on the next tool use, allowing for immediate testing and refinement of your guardrails.

Where are Hookify rules stored?

Rules are stored in your project's root within the .claude/ directory, following the naming convention .claude/hookify.{rule-name}.local.md.

Hookify Rule Architect

Name: Hookify Rule Architect
Author: kgadek

bykgadek

Developer Tools

Streamlines the creation of custom guardrails and automated triggers to govern Claude Code's behavior during terminal and file operations.

About

Hookify Rule Architect provides comprehensive guidance for implementing project-specific rules within Claude Code using the Hookify plugin. It enables users to define custom regex patterns that trigger warnings or blocks when specific bash commands are executed, files are modified, or tasks are completed. By leveraging YAML frontmatter and Markdown, this skill helps developers enforce security policies, prevent dangerous terminal commands, and maintain coding standards through automated, context-aware feedback directly within the AI workflow.

Key Features

Multi-condition logic support for granular rule triggering
Support for bash, file, stop, and prompt event types
0 GitHub stars
Customizable warning and block actions with Markdown messaging
Regex-based pattern matching for bash commands and file edits
Dynamic rule loading from the .claude/ directory

Use Cases

Enforcing security best practices by blocking hardcoded secrets in .env files
Preventing accidental execution of dangerous shell commands like rm -rf
Creating mandatory completion checklists that appear before the AI finishes a task

About

Key Features

Multi-condition logic support for granular rule triggering
Support for bash, file, stop, and prompt event types
0 GitHub stars
Customizable warning and block actions with Markdown messaging
Regex-based pattern matching for bash commands and file edits
Dynamic rule loading from the .claude/ directory

Use Cases

Enforcing security best practices by blocking hardcoded secrets in .env files
Preventing accidental execution of dangerous shell commands like rm -rf
Creating mandatory completion checklists that appear before the AI finishes a task