guard

Enforce destructive-command warnings and directory-based edit boundaries.

Updated Apr 12, 2026
One-click install
npx skills add https://github.com/singhianand23045/GarryTan --skill guard-singhianand23045
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/singhianand23045/GarryTan/tree/main/guard
Command: npx skills add https://github.com/singhianand23045/GarryTan --skill guard-singhianand23045

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Full safety mode prevents destructive commands and unintended edits by enforcing warning prompts and directory-based edit boundaries.

Core Features & Use Cases

  • Destructive command warnings prior to executing dangerous operations (rm -rf, etc.)
  • Directory-based edit boundary enforcement to block edits outside a specified path
  • Coordinated safety checks that integrate with sibling /careful and /freeze skills

Quick Start

Specify the directory to protect, then activate guard mode to enable destructive-command warnings and directory-scoped edit restrictions.

Frequently Asked Questions about guard

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

FAQPage Schema
How do I prevent destructive bash commands from running in production environments?

To prevent destructive bash commands in production, you need safety mechanisms that enforce warning prompts before executing dangerous operations like rm -rf. This functionality intercepts commands via PreToolUse hooks, applying checks that halt unintended edits and warn against dangerous operations.

How do I restrict file edits to a specific directory path during live system debugging?

Restricting file edits to a specific directory path requires directory-based edit boundary enforcement. This mechanism blocks any modifications attempted outside a defined safe path, ensuring live system debugging workflows remain confined to authorized areas without risking external files.

What are edit boundaries and how do they protect live systems during debugging?

Edit boundaries are defined directory paths that block unintended edits outside specified safe zones. They protect live systems during debugging by enforcing strict path limitations, ensuring modifications only occur within authorized directories while dangerous operations trigger warnings.

Can I coordinate multiple safety checks for DevOps workflows simultaneously?

Yes, you can coordinate multiple safety checks for DevOps workflows simultaneously. This functionality integrates with sibling careful and freeze scripts, combining destructive-command warnings and directory-based edit restrictions to enforce comprehensive production safety across interconnected operations.

Does this production safety tool require any external dependencies to function?

No, this production safety tool does not require any external dependencies to function. It operates independently using internal PreToolUse hooks to execute safety checks against destructive commands and edit boundaries without needing additional software packages or libraries.

When should I not rely on directory-based edit boundaries for production safety?

You should not rely on directory-based edit boundaries when your debugging workflow requires modifying files across multiple unrestricted system paths. This approach is specifically designed for confined edits within a defined path, meaning broad filesystem operations will be actively blocked.