guard

Enforce directory-boundary restrictions and destructive-command warnings during tool use.

Updated Mar 28, 2026
One-click install
npx skills add https://github.com/deveshkumars/Intentra --skill guard-deveshkumars
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/deveshkumars/Intentra/tree/main/guard
Command: npx skills add https://github.com/deveshkumars/Intentra --skill guard-deveshkumars

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Guard mode provides dual safety: it warns about destructive commands and enforces a directory-boundary to prevent unintended edits in critical systems.

Core Features & Use Cases

  • Combines destructive command warnings with a configurable edit boundary to protect production environments.
  • Lets teams quickly lock down directories while performing debugging or maintenance.
  • Useful for operations where accidental rm -rf or unauthorized edits must be prevented.

Quick Start

Ask the user which directory should be protected and enable guard mode to restrict edits to that path while enabling destructive-command warnings.

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 during production maintenance?

Preventing destructive bash commands during production maintenance requires enabling guard mode to enforce warnings and restrict edits to a specific directory boundary. This dual safety mechanism stops accidental execution of commands like rm -rf.

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

An edit boundary is a configurable directory restriction that prevents unauthorized edits to critical system paths. By locking down a specific directory, it ensures debugging or maintenance operations do not inadvertently modify protected production files.

How do I lock down a directory to restrict edits while debugging?

Locking down a directory to restrict edits while debugging requires specifying the target path to protect and enabling guard mode. This enforces the directory boundary and activates destructive command warnings simultaneously.

Do I need any other skills installed to use guard mode for workflow safety?

Using guard mode for workflow safety requires the careful and freeze skills to be installed and available. Guard mode leverages their hook scripts to enforce safety rules and directory boundaries.

Can I configure guard scripts to only warn about destructive commands without blocking edits?

Guard mode provides dual safety by default, combining destructive command warnings with a directory edit boundary. The metadata does not specify an option to disable the edit boundary while retaining only command warnings.