Where should I store my Hookify rules?

Rules should be saved in the .claude/ directory of your project using the specific naming convention: .claude/hookify.{rule-name}.local.md.

Can I prevent Claude from running certain commands?

Yes, by setting the action field to 'block' in the YAML frontmatter, you can prevent Claude from executing specific bash commands or completing operations that match your defined pattern.

What is a Hookify rule?

A Hookify rule is a markdown file with YAML frontmatter that allows you to define patterns to watch for and messages to display when those patterns are triggered during a Claude Code session.

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

No, Hookify rules are read dynamically. Any changes you make to a rule file take effect on the very next tool use within your active Claude Code session.

What events can trigger a Hookify rule?

Rules can trigger on bash commands, file operations (edits and writes), when the agent attempts to stop (completion checks), or when a user submits a prompt.

Hookify Rule Configuration

Name: Hookify Rule Configuration
Author: anthropics

byanthropics

•

291

개발자 도구

Guides the creation and management of custom guardrails and automation rules to control Claude's behavior during development.

소개

This skill provides a comprehensive framework for writing Hookify rules, which are project-specific guardrails stored as markdown files with YAML frontmatter. It enables developers to define patterns that trigger specific warnings or blocks when Claude attempts to run dangerous bash commands, edit sensitive files, or conclude a session without meeting requirements. By leveraging regex patterns and multi-condition logic, this skill ensures that project standards are maintained and risky operations are intercepted before they occur, making the AI's interactions with your codebase safer and more predictable.

주요 기능

Dynamic rule loading that updates behavior without restarting the session
Configurable actions to either warn the agent or block operations entirely
Comprehensive regex guidance for precise pattern matching in code and commands
291 GitHub stars
Multi-condition pattern matching for bash, file, and prompt events
Standardized YAML frontmatter generation for rule metadata

사용 사례

Flagging non-production code patterns such as console.log or hardcoded credentials
Enforcing mandatory pre-deployment checklists before an agent session stops
Preventing high-risk bash commands like recursive deletions or privilege escalation

소개

주요 기능

Dynamic rule loading that updates behavior without restarting the session
Configurable actions to either warn the agent or block operations entirely
Comprehensive regex guidance for precise pattern matching in code and commands
291 GitHub stars
Multi-condition pattern matching for bash, file, and prompt events
Standardized YAML frontmatter generation for rule metadata

사용 사례

Flagging non-production code patterns such as console.log or hardcoded credentials
Enforcing mandatory pre-deployment checklists before an agent session stops
Preventing high-risk bash commands like recursive deletions or privilege escalation