guard

Enforce directory-scoped edit restrictions and warn before destructive commands.

Updated Mar 21, 2026
One-click install
npx skills add https://github.com/nocnocgroup/gstack --skill guard-nocnocgroup
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/nocnocgroup/gstack/tree/main/guard
Command: npx skills add https://github.com/nocnocgroup/gstack --skill guard-nocnocgroup

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents dangerous changes by combining destructive-command warnings with directory-scoped edit restrictions. This ensures safe operation during maintenance or debugging, reducing the risk of accidental damage.

Core Features & Use Cases

  • Destructive command warnings: alerts before commands like rm -rf, DROP TABLE, or force-push are executed.
  • Edit boundary enforcement: restricts edits to a specified directory via a freeze boundary.
  • Combined safety: merges the capabilities of careful and freeze into a single safety-enhanced workflow for production or live debugging.

Quick Start

Enable guard mode before touching production by configuring the freeze boundary and using guarded commands for any risky edits.

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 like rm -rf during production maintenance?

You can prevent accidental destructive commands during production maintenance by enabling warnings that alert before commands like rm -rf, DROP TABLE, or force-push execute, reducing the risk of accidental damage in live environments.

How do I restrict file edits to a specific directory during live debugging?

You can restrict file edits to a specific directory during live debugging by setting a freeze boundary that enforces directory-scoped restrictions, ensuring only safe changes occur within the defined path.

Can I use Bash-based checks to warn before unsafe commands execute?

Yes, you can use Bash-based checks to warn before unsafe commands execute by running them before tool use, intercepting and preventing destructive actions during maintenance and debugging workflows.

What is the best way to combine destructive-command warnings with directory-scoped edit restrictions?

The best way to combine destructive-command warnings with directory-scoped edit restrictions is to merge them into a single safety-enhanced workflow that provides combined safety for production or live debugging.

Do I need integrated hooks to enable guard mode for production safety?

Yes, you need integrated hooks from careful and freeze to enable guard mode for production safety, providing the necessary Bash-based checks and freeze boundaries to lock down edits and warn before destructive actions.

When should I not use directory-scoped edit restrictions for maintenance?

You should avoid directory-scoped edit restrictions for maintenance when you need to make changes outside the defined freeze boundary, as the restrictions limit edits to a specified directory and block modifications outside that path.