skills.rest
steffenpharai
steffenpharaiCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

guard

Warn before destructive commands and block edits outside a configured directory.

Updated Aug 27, 2026
Search Tags:#safety#risk-mitigation#guard#destructive-commands#edit-protection#careful-integration#freeze-boundary
One-click install
npx skills add https://github.com/steffenpharai/StartupFramework --skill guard-steffenpharai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/steffenpharai/StartupFramework/tree/main/.claude/skills/gstack/guard
Command: npx skills add https://github.com/steffenpharai/StartupFramework --skill guard-steffenpharai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Full safety mode: destructive command warnings + directory-scoped edits. Combines /careful (warns before rm -rf, DROP TABLE, force-push, etc.) with /freeze (blocks edits outside a specified directory). Use for maximum safety when touching prod or debugging live systems. Use when asked to "guard mode", "full safety", "lock it down", or "maximum safety".

Core Features & Use Cases

  • Warnings before destructive commands across tooling, such as rm -rf, DROP TABLE, or force-push.
  • Directory-boundary enforcement that blocks edits outside the configured freeze directory.
  • Seamless integration with sibling skills to provide combined safety during maintenance.

Quick Start

Run guard mode to activate safety protections and set a freeze boundary for edits.

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 running accidentally?▼

To prevent destructive commands like rm -rf or DROP TABLE, you need a safety layer that warns before execution. Guard mode provides this by actively intercepting and warning against risky commands across various tooling during live system maintenance.

How can I restrict file edits to a specific directory during production debugging?▼

To restrict file edits to a specific directory during production debugging, you need directory-boundary enforcement. Guard mode blocks any edits attempted outside a configured freeze directory, ensuring risky changes are contained.

Do I need to install other safety skills to use full guard mode?▼

Yes, you need to install sibling safety skills to use full guard mode. Guard mode relies on the /careful and /freeze skills to function correctly, combining their destructive command warnings and directory-scoped edit restrictions.

When should I activate maximum safety mode for my production environment?▼

You should activate maximum safety mode when touching production or debugging live systems. Guard mode provides combined warnings for destructive commands and edit boundaries when asked to lock it down or ensure full safety.

What is the difference between guard mode and standard edit protection?▼

Guard mode offers full safety by combining destructive command warnings with directory-scoped edit blocking, whereas standard edit protection only restricts file modifications. Guard mode integrates multiple safety layers for maximum risk mitigation.

Can I block force-push operations while working on live systems?▼

Yes, you can block force-push operations while working on live systems. Guard mode warns before executing destructive commands across tooling, which includes force-push, DROP TABLE, and rm -rf.

Related Skills

Explore More Software Engineering Skills→
DicklesworthstoneDicklesworthstone
70

dcg

Block destructive Git and filesystem commands in Claude Code workflows.

Community
Advanced
EhsaanArkEhsaanArk

guard

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

Community
Advanced
akiyoshiakiyoshi

guard

Activate destructive command warnings and an edit boundary for protected directories.

Community
Advanced
managedcodemanagedcode
42

dotnet-mcaf-human-review-planning

Create structured manual review plans for large AI-generated codebases.

Official
Advanced