guard

Warn before destructive commands and restrict edits to a specified directory.

2|1|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/vnmoorthy/Skylog --skill guard-vnmoorthy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/vnmoorthy/Skylog/tree/main/.claude/skills/gstack/guard
Command: npx skills add https://github.com/vnmoorthy/Skylog --skill guard-vnmoorthy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Guard mode provides dual protections: it emits warnings for destructive commands and enforces a directory edit boundary to prevent accidental damage in prod or during debugging.

Core Features & Use Cases

  • Destructive command warnings: prompts before executing rm -rf, DROP TABLE, force-push, etc.
  • Edit boundary enforcement: restricts edits to a user-specified directory.
  • Seamless integration: combines two existing safety-focused skills (/careful and /freeze) into a single, installable workflow.
  • Use Case: You are about to perform a risky operation on a project; guard mode warns and confines edits to a safe sandbox.

Quick Start

Ask for the directory to restrict edits to and activate guard mode to enable destructive-command warnings and edit boundaries.

Frequently Asked Questions about guard

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

FAQPage Schema
How do I prevent accidental data loss from destructive bash commands during debugging?

To prevent accidental data loss during debugging, this safety mode emits warnings before executing destructive bash commands like rm -rf or DROP TABLE. It enforces pre-check hooks to verify boundaries and restricts edits to a user-specified directory.

How do I enforce a directory edit boundary to protect production environments?

You can enforce a directory edit boundary by activating guard mode and specifying the target directory. This restricts all file edits to that sandbox, preventing accidental modifications to production environments during maintenance or risky operations.

Do I need to install other components to enable destructive command warnings?

Yes, you need to install the optional careful and freeze components to fully enable destructive command warnings and edit boundaries. This Skill combines those two existing safety-focused skills into a single installable workflow.

What is the best way to add pre-check hooks for unsafe edits in a DevOps workflow?

The best way to add pre-check hooks in a DevOps workflow is using a safety mode that integrates boundary verification. It blocks unsafe edits and prompts before risky commands like force-push, ensuring coordinated safety across scripts.

Can I restrict file modifications to a specific sandbox when running risky operations?

Yes, you can restrict file modifications to a specific sandbox. When you activate this safety mode, it confines all edits to the directory you specify, warning before destructive commands and preventing accidental damage outside the boundary.

Why should I use a dedicated safety mode instead of manually checking destructive commands?

Using a dedicated safety mode automates boundary verification through pre-check hooks, preventing human error. It seamlessly combines warning prompts for destructive commands with edit boundary enforcement, providing coordinated protection that manual checking cannot reliably guarantee.