skills.rest
ComputerConnection
ComputerConnectionCOMMUNITY
Explore all skills by this author →
GitHub

guard

Warn on destructive commands and enforce per-directory edit boundaries.

Updated Mar 31, 2026
Search Tags:#devops#safety#freeze#boundary#guard#destructive-warnings
One-click install
npx skills add https://github.com/ComputerConnection/zach-pack --skill guard-computerconnection
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/ComputerConnection/zach-pack/tree/main/skills/gstack-guard
Command: npx skills add https://github.com/ComputerConnection/zach-pack --skill guard-computerconnection

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents accidental system damage by warning about destructive commands and blocks edits outside a defined directory boundary.

Core Features & Use Cases

  • Destructive-command warnings before executing dangerous operations (rm -rf, DROP TABLE, etc.).
  • Directory-scoped edit boundaries that prevent changes outside a specified path.
  • Quick integration with existing safety hooks to enforce safeguards during live deployments and debugging.

Quick Start

Enable guard mode and specify the directory to protect; the system will warn on dangerous commands and enforce 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 destructive commands during production deployments?▼

Directory-scoped edit boundaries prevent file modifications outside a specified path by validating operations against the defined boundary. This mechanism blocks unauthorized edits during debugging or refactoring, ensuring changes remain isolated to the approved directory scope.

Can I enforce safe editing boundaries when debugging in a live development environment?▼

Destructive-command warnings work by implementing pre-use hooks that validate commands against known dangerous patterns before execution. When operations like rm -rf or DROP TABLE are detected, the system issues a warning to prevent accidental system damage during maintenance.

Does this safety skill coordinate with other components to enforce freeze states?▼

Yes, the safety skill coordinates with sibling careful and freeze components to ensure safeguards remain active. This coordination enforces comprehensive boundary controls and destructive-command warnings across the workflow during deployments and refactoring tasks.

What are the limitations of relying on pre-use hooks for command validation?▼

Pre-use hooks for command validation depend on recognizing known dangerous patterns, meaning novel or obfuscated destructive commands might bypass detection. The system focuses on warning about specified threats and enforcing directory boundaries, requiring explicit configuration for comprehensive protection.

Related Skills

Explore More Software Engineering Skills→
agwacomagwacom

guard

Restrict edits to a chosen directory and warn before destructive commands.

Official
Advanced
scanbottscanbott

unfreeze

Remove the edit restriction set by /freeze across all project directories.

Community
Basic
SakshamKandelSakshamKandel

guard

Warn and block destructive commands and edits outside a safe directory boundary.

Community
Advanced
BaGyun0107BaGyun0107

guard

Warn before destructive bash commands and restrict file edits to safe directories.

Community
Advanced