engineer-hook

Manage IDE-specific event hooks for Cursor, Claude Code, and OpenCode.

Updated May 16, 2026
One-click install
npx skills add https://github.com/korchasa/flowai-plugins --skill engineer-hook
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: engineer-hook
Source: https://github.com/korchasa/flowai-plugins/tree/main/plugins/flowai-devtools/skills/engineer-hook
Command: npx skills add https://github.com/korchasa/flowai-plugins --skill engineer-hook

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires cursor, claude-code, opencode, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

The engineer-hook Skill unit addresses the need for managing event hooks/plugins that control agent behavior, command filtering, and automation. It simplifies the process of creating, configuring, and implementing hooks across different Integrated Development Environments (IDEs).

Core Features & Use Cases

  • IDE-Specific Hook Management: Supports Cursor, Claude Code, and OpenCode IDEs with specific hook configurations for each.
  • Event Hooks Creation: Allows users to create hooks that intercept agent actions and apply rules like allowing, blocking, or modifying actions.
  • Command Filtering and Automation: Provides tools for auditing, security checks, and automation tasks based on hook rules.
  • Use Case: For a software engineer who needs to implement a security check for all commands executed in an IDE, the engineer-hook Skill can be used to create a custom hook that filters out dangerous commands.

Quick Start

Run the engineer-hook skill to create a new hook for command filtering in Cursor IDE.

Frequently Asked Questions about engineer-hook

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create event hooks to filter dangerous commands executed by AI agents in my IDE?

You can create event hooks that intercept agent actions to allow, block, or modify commands based on custom rules. This Skill manages command filtering by requiring IDE-specific configuration files and script execution for Cursor, Claude Code, and OpenCode.

Can I use event hooks to request user confirmation before an AI agent executes a command in Cursor?

Yes, event hooks can request confirmation for agent actions in Cursor. You create custom logic to intercept agent behavior and enforce rules that require manual approval before commands execute, managed through IDE-specific configuration files.

Does this command filtering approach work across different IDEs like Claude Code and OpenCode?

Command filtering works across Cursor, Claude Code, and OpenCode. The Skill manages IDE-specific hook configurations for each environment, allowing custom automation and security rules to intercept and modify agent actions consistently.

What is the best way to implement security checks for all commands executed by an IDE agent?

Implement security checks by creating custom event hooks that audit and filter commands. This Skill allows you to define rules that block dangerous agent actions and automate security tasks across Cursor, Claude Code, and OpenCode.

Do I need specific configuration files to set up agent behavior automation in OpenCode?

Yes, setting up agent behavior automation requires IDE-specific configuration files and script execution. The Skill uses these files to deploy event hooks that control command filtering and automation logic for OpenCode.

Why use event hooks instead of standard IDE settings for managing AI agent behavior?

Event hooks provide custom logic to block, modify, or request confirmation for agent actions, which standard IDE settings cannot enforce. This Skill applies advanced filtering and automation rules across Cursor, Claude Code, and OpenCode.