guard

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

Updated Feb 28, 2026
One-click install
npx skills add https://github.com/Analog-Labs/pixelport-launchpad --skill guard-analog-labs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/Analog-Labs/pixelport-launchpad/tree/main/.agents/skills/gstack/guard
Command: npx skills add https://github.com/Analog-Labs/pixelport-launchpad --skill guard-analog-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides a unified safety layer that issues warnings for destructive commands (like rm -rf, DROP TABLE) and enforces a directory-based edit boundary to prevent risky changes in production or sensitive projects.

Core Features & Use Cases

  • Destructive command warnings: integrates with /careful to risk-check dangerous commands before execution.
  • Edit boundary enforcement: uses /freeze to restrict edits to a user-specified directory.
  • Protection in production/debugging: reduces risk when performing live changes or debugging.

Quick Start

Provide a directory path to restrict edits and enable destructive-command warnings to begin guard mode.

Frequently Asked Questions about guard

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

FAQPage Schema
How do I prevent destructive commands like rm -rf from running in production?

Preventing destructive commands in production requires a safety layer that risk-checks commands like rm -rf or DROP TABLE before execution. This approach confines edits to a safe directory and issues warnings before dangerous commands run.

What is a directory-scoped edit boundary and how does it protect live environments?

A directory-scoped edit boundary restricts file modifications to a user-specified safe directory. It protects live environments by ensuring that debugging or production changes cannot accidentally alter files outside the defined boundary.

How do I set up guard mode to restrict edits to a specific directory?

Setting up guard mode involves providing a directory path to restrict edits and enabling destructive-command warnings. This activates a safety layer combining edit boundaries with warnings for risky commands.

Can I apply edit boundaries and command warnings when debugging remote environments?

Yes, you can apply edit boundaries and command warnings to remote environments. The safety layer works across local or remote setups, reducing risk when performing live changes or debugging.

What are the limitations of using directory-based edit boundaries for production protection?

Directory-based edit boundaries limit modifications to a specific safe directory but do not block the execution of dangerous commands themselves. They must be paired with destructive command warnings to achieve full production protection.