guard

Activate destructive-command warnings and directory-scoped edit restrictions.

Updated Apr 11, 2026
One-click install
npx skills add https://github.com/Scivor/helm --skill guard-scivor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/Scivor/helm/tree/main/skills/gstack/guard
Command: npx skills add https://github.com/Scivor/helm --skill guard-scivor

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Activates both destructive command warnings and directory-scoped edit restrictions to protect critical systems during development and debugging.

Core Features & Use Cases

  • Combines /careful and /freeze safety features into a single activation for maximum protection.
  • Enforces a user-specified edit boundary and warns before destructive commands.
  • Suitable for production touches, debugging live systems, or any scenario requiring strict operational guardrails.

Quick Start

Ask the user which directory to restrict edits to, then enable guard mode to apply destructive command warnings and a directory-wide edit freeze.

Frequently Asked Questions about guard

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

FAQPage Schema
How do I prevent accidental destructive commands when debugging live production systems?

Preventing destructive commands in production requires activating a guard mode that enforces directory-scoped edit restrictions and triggers warnings before harmful operations execute. It protects critical systems during live debugging.

How do I set up directory-scoped edit restrictions for collaborative development?

Directory-scoped edit restrictions are set up by specifying a target directory to restrict edits to, then enabling guard mode to apply a directory-wide edit freeze. This enforces strict edit boundaries for collaborative environments.

Do I need any specific extensions installed to enforce edit boundaries and command warnings?

Yes, enforcing edit boundaries and command warnings requires the /careful and /freeze skills to be installed beforehand. The system uses these dependencies to store the freeze boundary in a state file and report the active guard status.

What is the best way to freeze edits and warn against destructive operations simultaneously?

The best way to freeze edits and warn against destructive operations is to combine both safety features into a single activation. This mode applies a directory-wide edit freeze alongside destructive command warnings for maximum protection.

Can I use guard mode to protect critical systems during DevOps operations?

Yes, you can use guard mode to protect critical systems during DevOps operations by enforcing a user-specified edit boundary and warning before destructive commands. It is suitable for production touches and strict operational guardrails.