How do I trigger a rule only for specific file types?

You can use advanced conditions with the 'file_path' field and a 'regex_match' operator to target specific extensions like '.env' or '.tsx'.

What are Hookify rules in Claude Code?

Hookify rules are configuration files that define patterns Claude should watch for. When a pattern matches a command or file edit, Claude triggers a predefined message or action.

Can Hookify rules block specific actions?

Yes, by setting the action field to 'block' in the YAML frontmatter, you can prevent specific bash commands from running or stop problematic file edits.

Do I need to restart Claude to apply new rules?

No, rules are read dynamically and changes take effect on the very next tool use after the rule file is saved.

Where should I store my Hookify rule files?

Rules should be stored in the `.claude/` directory of your project root with the naming convention `.claude/hookify.{rule-name}.local.md`.

Hookify Rule Architect

Name: Hookify Rule Architect
Author: Activer007

byActiver007

0•

开发者工具

Streamlines the creation and management of Hookify rules to enforce coding standards, security guardrails, and automated checklists in Claude Code.

The Hookify Rule Architect skill provides a structured framework for defining event-driven triggers within the Claude Code environment. By utilizing markdown files with YAML frontmatter, it allows developers to configure Claude to monitor bash commands, file modifications, and session completion events. This skill is essential for implementing project-specific best practices—such as preventing destructive shell commands, flagging debug logs before commits, or ensuring documentation is updated—through a robust system of regex patterns and customized markdown feedback.

主要功能

01Dynamic rule loading from project-local directories for instant workflow enforcement.

02Advanced conditional logic with regex matching, substring checks, and multi-field validation.

030 GitHub stars

04Standardized markdown-based rule definitions with YAML frontmatter.

05Customizable response actions to either warn the agent or block specific operations.

06Support for multiple event types including bash, file, stop, and prompt hooks.

使用场景

01Implementing mandatory completion checklists that trigger when the agent attempts to stop a session.

02Preventing accidental execution of dangerous commands like 'rm -rf' or unauthorized 'sudo' usage.

03Enforcing code quality by automatically flagging 'console.log', 'eval', or leaked secrets in file edits.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add activer007/ordinary-claude-skills rule-identifier

For use in Claude.ai and ChatGPT

Download Skill