guard

Blocks edits outside a designated directory and warns about destructive commands in Bash hooks.

1|Updated Mar 23, 2026
One-click install
npx skills add https://github.com/txema-puch/drone-ai-saturdays --skill guard-txema-puch
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/txema-puch/drone-ai-saturdays/tree/main/.claude/skills/gstack/guard
Command: npx skills add https://github.com/txema-puch/drone-ai-saturdays --skill guard-txema-puch

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents accidental destructive operations and uncontrolled edits in critical environments by combining two safety mechanisms: destructive-command warnings and directory-scoped edit restrictions.

Core Features & Use Cases

  • Destructive command warnings for rm -rf, DROP TABLE, force-push, and similar operations.
  • Directory-scoped edit boundary enforcement to restrict edits to a specified path.
  • Seamless composition of both capabilities for production safety during maintenance and debugging.
  • Use case: DevOps engineers enabling guard mode before touching production or sensitive systems.

Quick Start

Enable guard mode before making any edits and confirm destructive-command warnings are active.

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 like rm -rf or DROP TABLE during production maintenance?

To prevent destructive commands like rm -rf or DROP TABLE during production maintenance, you can enable safety boundaries that issue warnings before executing dangerous operations. This guard mode targets DevOps workflows to prevent accidental outages and data loss.

Can I restrict file edits to a specific directory during debugging workflows?

Yes, you can restrict file edits to a specific directory during debugging workflows by applying directory-scoped edit boundaries. This confines modifications to a designated path, preventing uncontrolled mutations in critical environments.

What is the best way to enforce safety guardrails for force-push and other destructive operations in Bash?

The best way to enforce safety guardrails for force-push and destructive operations in Bash is integrating hooks that combine destructive-command warnings with edit restrictions. This composition provides full safety mode for production maintenance and debugging.

How do directory-scoped edit boundaries work with destructive-command warnings?

Directory-scoped edit boundaries work with destructive-command warnings by composing both mechanisms into a full safety mode. The warnings intercept dangerous commands while the boundary blocks edits attempted outside a specified directory path.

Do I need Bash-based hooks to enable guard mode for sensitive systems?

Yes, you need Bash-based hooks to enable guard mode for sensitive systems, as the safety mechanisms integrate directly with them. This setup ensures warnings and boundary restrictions are actively enforced before any commands execute.