guard

Enforce edit boundaries and warn on destructive commands during debugging.

Updated Mar 11, 2026
One-click install
npx skills add https://github.com/jkhowland/jkhowland --skill guard-jkhowland
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/jkhowland/jkhowland/tree/main/.claude/skills/gstack/guard
Command: npx skills add https://github.com/jkhowland/jkhowland --skill guard-jkhowland

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides a unified safety layer that warns before destructive commands and restricts edits to a user-specified directory, reducing the risk of accidental production damage during maintenance or debugging.

Core Features & Use Cases

  • Destructive-command warnings: Warns before risky operations such as rm -rf, DROP TABLE, or force-push attempts during active work on prod systems.
  • Edit boundary enforcement: Blocks edits outside a user-selected directory to prevent unintended changes to critical assets.
  • Pre-tool gating with careful and freeze: Integrates with sibling /careful and /freeze scripts to validate and enforce safety boundaries during tool use.
  • Use cases: Ideal for live debugging, prod maintenance, or any scenario requiring "guard mode" or maximum safety.

Quick Start

Ask the user which directory to restrict edits to, then activate guard mode to enable warnings and the edit boundary.

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 commands during live production debugging?

Prevent accidental destructive commands during production debugging by activating a safety layer that warns before risky operations like rm -rf or DROP TABLE and restricts edits to a user-specified directory.

Can I restrict file edits to a specific directory boundary during maintenance?

Yes, you can restrict file edits to a specific directory boundary during maintenance by specifying your target path, which blocks any unintended changes to critical assets outside that chosen directory.

What is an edit boundary and how does it protect production systems?

An edit boundary protects production systems by enforcing a safe operating perimeter around a chosen directory, validating tool use through pre-tool hooks to block modifications outside that path.

Does this safety layer warn before force-push attempts and database drops?

Yes, this safety layer warns before force-push attempts and database drops by intercepting destructive commands during active work on production systems to reduce the risk of accidental damage.

How do I start using guard mode for maximum production safety?

Start using guard mode for maximum production safety by specifying which directory to restrict edits to, which activates the warnings and edit boundary enforcement for your maintenance session.

When should I not rely on an automated edit boundary for prod changes?

You should not rely solely on an automated edit boundary when performing broad system migrations outside the chosen path, as the safety layer specifically restricts edits to a single user-specified directory.