guard

Warn on destructive commands and restrict edits to a chosen directory.

1|Updated May 4, 2026
One-click install
npx skills add https://github.com/sanskarbasnet/fstack --skill guard-sanskarbasnet
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/sanskarbasnet/fstack/tree/main/guard
Command: npx skills add https://github.com/sanskarbasnet/fstack --skill guard-sanskarbasnet

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents destructive actions and enforces a secure editing boundary, reducing risk during maintenance and debugging.

Core Features & Use Cases

  • Warnings for destructive commands (rm -rf, DROP TABLE, force-push, etc.) to prevent accidental damage.
  • Directory-bound edits via a freeze boundary to block edits outside a chosen path.
  • Pre-tool safety checks and coordinated brain-sync to keep teammates aligned during risky operations.

Quick Start

Ask the user to specify the directory to protect, then enable guard mode with the /careful and /freeze components.

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 or DROP TABLE during production maintenance?

To prevent accidental destructive commands during production maintenance, you can enforce safety mode with warnings for destructive commands like rm -rf or DROP TABLE. This blocks unsafe actions before execution, reducing risk during live-debug sessions.

What is the best way to restrict file edits to a specific directory boundary?

The best way to restrict file edits to a specific directory boundary is by applying a freeze boundary. This enforces directory-bound edits, blocking any modifications outside your chosen protected path during risky operations.

How do I coordinate with teammates to prevent unsafe edits during live-debug sessions?

You coordinate with teammates to prevent unsafe edits during live-debug sessions using pre-tool safety checks and brain-sync. This synchronizes shared safety states, keeping aligned during risky operations.

Can I use directory-bound edits and destructive-command warnings together for prod maintenance?

Yes, you can use directory-bound edits and destructive-command warnings together for prod maintenance. Enabling full safety mode combines both safeguards, preventing accidental damage and unsafe edits simultaneously.

How do I enable full safety mode to block force-push and other unsafe actions?

To enable full safety mode and block unsafe actions like force-push, specify the directory to protect and activate the guard components. This applies warnings and edit boundaries to prevent accidental damage.