guard

Enforce destructive-command warnings and directory-bound edits for safety.

Updated Apr 6, 2025
One-click install
npx skills add https://github.com/yolopunk/jedi --skill guard-yolopunk
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/yolopunk/jedi/tree/main/.claude/skills/gstack/guard
Command: npx skills add https://github.com/yolopunk/jedi --skill guard-yolopunk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Activates safety protections by providing destructive-command warnings and by enforcing a directory boundary to prevent risky edits in production environments.

Core Features & Use Cases

  • Combines destructive-command warnings with directory-scoped edits to prevent accidental damage.
  • Integrates with /careful and /freeze to deliver a unified safety workflow across shell actions and file edits.
  • Suitable for development, staging, and production touchpoints where risk mitigation is critical.

Quick Start

Activate guard mode to enable destructive-command warnings and restrict edits to a chosen directory.

Frequently Asked Questions about guard

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent accidental file edits outside a specific directory?

Directory-bound edits prevent accidental damage by restricting file modifications to a chosen path. This safety mechanism blocks edit and write operations outside the designated boundary, protecting sensitive production systems from unintended changes.

What are destructive-command warnings and how do they work?

Destructive-command warnings trigger alerts before executing risky shell actions that could cause irreversible damage. They intercept hazardous Bash commands using PreToolUse hooks, prompting confirmation to prevent accidental data loss in production or staging environments.

Can I apply directory boundary restrictions to production environments?

Yes, directory boundary restrictions are designed specifically for production and sensitive systems. The combined guard mode enforces directory-bound edits and destructive-command warnings simultaneously, mitigating risk for developers and operators working in critical environments.

How do I enable safety warnings for shell and file editing workflows?

Safety warnings for shell and file editing workflows are enabled by activating guard mode. This triggers PreToolUse hooks that intercept Bash, Edit, and Write operations, applying destructive-command warnings and directory-bound restrictions across the unified workflow.

What's the best way to combine destructive-command warnings with directory-scoped edits?

Combining destructive-command warnings with directory-scoped edits is best achieved through a unified guard mode. This integrated approach intercepts both shell actions and file modifications, delivering comprehensive risk mitigation across development, staging, and production touchpoints.