guard

Name: guard
Availability: InStock
Author: ehsaanark

Enable guard mode to warn before destructive commands and restrict edits to a directory.

Updated Mar 14, 2026

Search Tags:#safety #dev-tools #guard #destructive-commands #pre-checks #edit-boundary

Ready to run

npx skills add https://github.com/EhsaanArk/ai-signal-router --skill guard-ehsaanark

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/EhsaanArk/ai-signal-router/tree/main/.claude/skills/gstack/guard
Command: npx skills add https://github.com/EhsaanArk/ai-signal-router --skill guard-ehsaanark

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides a unified guard mode that prevents destructive commands and enforces an edit boundary, reducing the risk of accidental damage in development or production workflows.

Core Features & Use Cases

Manual trigger via /guard to activate safety protections.
Combines the safety behaviors of the /careful and /freeze skills to warn before dangerous actions and restrict edits to a specified directory.
Suitable for production work or live debugging where changes must be guarded and auditable.
Depends on sibling skill scripts for pre-checks before tool usage.

Quick Start

Trigger guard mode by typing /guard to enable warnings and a directory-bound edit restriction.

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 during production debugging?▼

Preventing destructive commands during production debugging requires enabling a guard mode that triggers pre-tool warnings before dangerous actions are executed. This Skill activates that mode manually, combining warning behaviors and directory-bound edit restrictions to protect live workflows.

What is the best way to enforce an edit boundary in a specific directory?▼

Enforcing an edit boundary in a specific directory requires activating a freeze mode that restricts file modifications to a designated path. This Skill combines boundary checks with destructive command warnings to ensure edits remain confined and auditable.

How do you set up full safety mode for live code edits?▼

Setting up full safety mode for live code edits involves triggering a unified guard mode that restricts directory boundaries and warns against dangerous commands. This Skill relies on sibling pre-check installations to execute safety validations before any tool use.

Do I need any dependencies installed to enforce edit boundaries and pre-checks?▼

Enforcing edit boundaries and pre-checks requires the careful and freeze skill installations to be present in the environment. This Skill relies on those sibling scripts to execute the necessary pre-tool hooks before any modifying actions occur.

When should I not use automated guard mode for development workflows?▼

Automated guard mode is not suited for development workflows where rapid unrestricted edits are required, as it strictly enforces edit boundaries and triggers warnings. It is designed for production or live debugging scenarios where changes must be guarded and auditable.

Can I restrict file modifications to a specific directory during live debugging?▼

Restricting file modifications to a specific directory during live debugging is achieved by activating a freeze boundary combined with destructive command warnings. This Skill enforces those directory-bound restrictions through pre-tool hooks to prevent accidental damage.