Can I block specific terminal commands?

Yes, by setting the action to 'block' in the rule's YAML frontmatter, you can prevent Claude from executing high-risk commands like 'rm -rf' or 'sudo'.

Do I need to restart Claude Code for new rules to take effect?

No, rules are read dynamically from the .claude directory, meaning any changes or new rules apply immediately without requiring a restart.

What are Claude Code Hookify rules?

Hookify rules are custom configuration files that allow you to monitor and intercept Claude's actions, such as running bash commands or editing files, based on specific patterns you define.

How do I target rules to specific file types?

You can use the advanced 'conditions' field to target specific file paths using regex, allowing you to create rules that only apply to files like '.env', '.tsx', or '.js'.

What event types can I monitor?

You can hook into bash commands, file edits (Write/Edit tools), prompt submissions, agent stop events, or use 'all' for universal monitoring.

Hookify Rule Architect

Name: Hookify Rule Architect
Author: Aventerica89

byAventerica89

0•

Security & Testing

Enables developers to create custom guardrails and automation rules for Claude Code by defining patterns that trigger specific warnings or actions.

The Hookify Writing Rules skill provides a comprehensive framework for building domain-specific constraints and safety nets within Claude Code. By using Markdown files with YAML frontmatter, users can define sophisticated rules that monitor terminal commands, file modifications, and user prompts. Whether you need to prevent accidental commits of API keys, block dangerous shell commands like 'rm -rf', or enforce specific coding standards through regex-based patterns, this skill offers the structured guidance and logic necessary to implement robust automated interventions in your AI-assisted development workflow.

Key Features

01Support for multiple event hooks including bash, file, stop, and prompt

02Advanced multi-condition logic using operators like regex_match, contains, and starts_with

03Custom regex-based pattern matching for shell commands and file edits

040 GitHub stars

05Flexible action responses allowing for either non-intrusive warnings or strict command blocking

06Seamless integration with .claude local configuration for immediate rule application

Use Cases

01Enforcing mandatory testing or linting steps before the AI agent concludes a session

02Preventing sensitive information like API keys from being saved in .env files

03Blocking high-risk shell commands like 'rm -rf' or insecure 'chmod' operations

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add aventerica89/claude-codex hookify-writing-rules

For use in Claude.ai and ChatGPT

Download Skill