What are Hookify rules in Claude Code?

Hookify rules are local markdown files with YAML frontmatter that define specific patterns for Claude to watch. When a pattern is matched during a bash command or file edit, Claude displays a predefined message or blocks the action.

Can Hookify rules prevent Claude from making mistakes?

Yes, by using the 'block' action in the frontmatter, you can stop Claude from executing dangerous commands or writing problematic code patterns before they are applied.

Which events can trigger a Hookify rule?

Rules can be triggered by bash commands (bash), file operations (file), when the agent wants to stop (stop), or when a user submits a new prompt (prompt).

Where should I store my Hookify rules?

Rules should be stored in the `.claude/` directory of your project using the naming convention `hookify.{rule-name}.local.md`. It is recommended to add these to your .gitignore.

Hookify Rule Assistant

Name: Hookify Rule Assistant
Author: opensesh

byopensesh

•

Productivity & Workflow

Defines custom guardrails and automated triggers to guide Claude's behavior and enforce project standards during development tasks.

The Hookify Rule Assistant skill provides comprehensive guidance for creating and managing project-specific rules within the Hookify framework for Claude Code. By utilizing markdown files with YAML frontmatter, it enables developers to intercept bash commands, file edits, and tool stops to provide immediate feedback, prevent dangerous operations, or enforce specific coding patterns. This skill is essential for teams looking to standardize AI-assisted workflows, automate code reviews, and maintain high security and quality standards through context-aware warnings and block actions.

Key Features

01Dynamic rule loading for real-time testing and refinement without restarts

02Advanced conditional logic using field-specific operators like regex_match and contains

03Syntax guidance for YAML frontmatter and regex pattern matching

041 GitHub stars

05Multi-event support including bash commands, file edits, and completion stops

06Template-based creation of informative warning and blocking messages

Use Cases

01Preventing the execution of destructive terminal commands like 'rm -rf' or unauthorized privilege escalation

02Enforcing security best practices by blocking sensitive file edits or detecting 'eval' and 'innerHTML' usage

03Creating automated completion checklists to ensure tests are run and documentation is updated before a task ends

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add opensesh/bos-2.0 writing-rules

For use in Claude.ai and ChatGPT

Download Skill