guard

Warn on destructive commands and restrict edits to a specified directory.

Updated Mar 19, 2026

Search Tags:#devops #production #bash #safety #destructive-warnings #edit-boundary #guard-mode

Ready to run

npx skills add https://github.com/onlypfachi/chief --skill guard-onlypfachi

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/onlypfachi/chief/tree/main/guard
Command: npx skills add https://github.com/onlypfachi/chief --skill guard-onlypfachi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Guards against dangerous operations by combining destructive command warnings with a boundary that restricts edits to a specific directory.

Core Features & Use Cases

Combines /careful and /freeze to provide both destructive command warnings and edit-boundary enforcement.
Suitable for production debugging and live systems where safety is critical.
Use when you need to prevent accidental data loss from destructive commands.

Quick Start

Specify a directory to restrict edits to and enable destructive-command warnings to activate guard mode.

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 bash commands in production?▼

You can prevent accidental destructive bash commands in production by enabling guard mode, which intercepts and warns against dangerous operations before execution. This protects live systems from unintended data loss during DevOps workflows.

Can I restrict file edits to a specific directory during live development?▼

Yes, you can restrict file edits to a specific directory by setting a directory path boundary. This edit-boundary enforcement ensures that modifications are confined to the approved scope, preventing accidental changes to critical files outside the target directory.

What is guard mode for maximum safety in DevOps workflows?▼

Guard mode for maximum safety in DevOps workflows is a combined protective state that integrates destructive command warnings with directory-scoped edit restrictions. It satisfies safety hooks to lock down live systems and prevent accidental destructive actions.

Does this safety tool require any dependencies to enforce a lock-down?▼

No, this safety tool requires no external dependencies to enforce a lock-down. It operates independently to provide full safety by combining cross-skill dependencies like careful and freeze for destructive-command warnings and edit-boundary enforcement.

When do I need to enable full safety warnings for production debugging?▼

You need to enable full safety warnings for production debugging when working directly on live systems where accidental data loss is a risk. Activating guard mode ensures destructive commands are intercepted before causing irreversible damage.