guard

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

1|Updated Mar 13, 2026
One-click install
npx skills add https://github.com/cattboy/lil_bro --skill guard-cattboy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/cattboy/lil_bro/tree/main/.claude/skills/guard
Command: npx skills add https://github.com/cattboy/lil_bro --skill guard-cattboy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides full safety by issuing destructive-command warnings and enforcing a directory-scoped edit boundary to prevent accidental changes to critical systems during development or debugging.

Core Features & Use Cases

  • Destructive-command warnings: prompts before executing dangerous commands like rm -rf, DROP TABLE, or force-push.
  • Edit-boundary enforcement: restricts edits to a user-specified directory, blocking changes outside that path.
  • Safe collaboration: can be used alongside other safety modules (/careful and /freeze) to create a robust safety envelope for prod or live debugging.

Quick Start

Ask for the directory to protect and enable guard mode, then run the setup to activate destructive-command warnings and the edit boundary.

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 during production debugging?

Preventing accidental destructive commands during production debugging requires issuing warnings before executing dangerous commands like rm -rf or DROP TABLE. This enables guard mode to prompt for confirmation, blocking high-risk edits.

What is an edit boundary and how does it protect critical systems?

An edit boundary restricts file modifications to a user-specified directory, blocking any changes outside that protected path. This prevents accidental edits to critical systems during development or debugging sessions.

How do I set up a directory-scoped edit boundary for development?

Setting up a directory-scoped edit boundary involves specifying the directory to protect and enabling guard mode. Running the setup activates destructive-command warnings and enforces the edit boundary for that path.

Does guard mode work with other safety modules for live debugging?

Guard mode works with other safety modules, coordinating with sibling safety modules to create a robust safety envelope. This combination ensures comprehensive protection during live or production debugging.

What's the best way to block force-push and similar dangerous operations?

The best way to block dangerous operations like force-push is to enable a safety-focused tool that issues destructive-command warnings. This intercepts high-risk commands before execution, preventing accidental data loss.