guard

Enforce safety checks on destructive commands and edit boundaries.

5|Updated Apr 24, 2026

Search Tags:#safety #risk-mitigation #live-debugging #destructive-commands #production-security #guard-mode #edit-boundaries

Ready to run

npx skills add https://github.com/timurgaleev/vibestack --skill guard-timurgaleev

💬 Prompt for ChatGPT / Claude Chat▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Full safety mode provides warnings for destructive commands and enforces a directory-boundary to prevent risky edits, especially when touching prod or debugging live systems.

Core Features & Use Cases

Destructive command warnings (e.g., rm -rf, DROP TABLE, force-push) are always on.
Edit boundary enforcement restricts edits to a user-specified directory.
Works in tandem with /careful and /freeze to provide end-to-end safety.

Quick Start

Set a freeze boundary for edits, then type /guard to activate combined safety protections.

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 like rm -rf or DROP TABLE from executing during live debugging?▼

Destructive command warnings are always on to prevent risky commands like rm -rf or DROP TABLE from executing. The guard mode intercepts these dangerous actions during live debugging and local sessions, enforcing safety checks before any damage occurs.

Can I restrict file edits to a specific directory boundary during production debugging?▼

Edit boundary enforcement restricts file edits to a user-specified directory boundary during production debugging. This prevents unauthorized edits and risky modifications from affecting critical systems outside the designated safe zone.

What is the best way to activate guard mode for combined safety protections?▼

To activate guard mode for combined safety protections, set a freeze boundary for edits, then type /guard. This works in tandem with /careful and /freeze integrations to provide end-to-end safety across your local sessions and live debugging scenarios.

Does guard mode work without additional dependencies or components?▼

Guard mode operates without additional dependencies or components, relying solely on its internal safety check logic. It enforces guard rules through the /careful and /freeze integrations and stores the freeze state in a deterministic location.

Why does force-push trigger a warning during a local session?▼

Force-push triggers a warning during a local session because destructive command warnings are always on. The system intercepts force-push and similar risky operations to enforce safety checks before they can impact your repository.

When should I not use edit boundary enforcement?▼

Edit boundary enforcement should not be used when you need to make edits across multiple directories outside the user-specified boundary. The restriction prevents any modifications outside the defined safe zone, which limits cross-directory workflows.