careful

Warn before executing destructive shell commands like rm -rf and kubectl delete.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides guardrails that warn before executing destructive commands (e.g., rm -rf, git reset --hard, kubectl delete) to prevent accidental data loss.

Core Features & Use Cases

  • Real-time warnings: Detects dangerous commands before they run and prompts for confirmation.
  • Safe overrides: Allows explicit user override with deliberate consent and audit logging.
  • Session logging: Records attempt details to ~/.gstack/analytics for auditing.

Quick Start

Enable careful mode and then perform a destructive command to receive a warning prompt before proceeding.

Frequently Asked Questions about careful

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

FAQPage Schema
How do I add a safety prompt before running destructive bash commands like rm -rf?

To add a safety prompt before destructive bash commands, this skill uses deterministic pre-checks via a Bash hook to intercept hazardous operations like rm -rf and warn you before execution. It requires explicit user consent to proceed with the action.

Can I override the warning prompt if I am sure I want to execute git reset --hard?

You can override the warning prompt for git reset --hard using a deliberate opt-out override mechanism. This safe override requires explicit consent and logs the usage data to an analytics file for auditing purposes.

How does audit logging work for destructive shell commands in local environments?

Audit logging for destructive shell commands records attempt details into the ~/.gstack/analytics directory. This session logging captures warnings and overrides to track hazardous operations across both local environments and production workflows.

Does this safety hook support common hazardous operations like kubectl delete?

This safety hook supports common hazardous operations including kubectl delete, git reset --hard, and rm -rf. It applies deterministic pre-checks across local environments and production workflows to prevent unintended data loss.

What is the best way to prevent unintended data loss from accidental shell commands?

The best way to prevent unintended data loss is to implement real-time guardrails that detect dangerous commands before they run. This skill identifies hazardous shell operations and applies configurable warnings to stop accidental execution.