careful

Warn before executing destructive Bash commands like rm -rf and DROP TABLE.

Updated Apr 15, 2026
One-click install
npx skills add https://github.com/256javy/safia --skill careful-256javy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/256javy/safia/tree/main/.claude/skills/gstack/careful
Command: npx skills add https://github.com/256javy/safia --skill careful-256javy

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python3.

What problem does it solve?

Prevents accidental damage by warning before executing destructive commands in Bash and related tooling, reducing the risk of data loss on prod or in shared environments.

Core Features & Use Cases

  • Warnings for dangerous commands such as rm -rf, git reset --hard, kubectl delete, and DROP TABLE.
  • Per-command override to allow users to proceed when appropriate.
  • Safe for use in production debugging and collaborative environments; supports optional suppression on trusted tasks.

Quick Start

Enable careful mode before running any potentially destructive command.

Frequently Asked Questions about careful

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

FAQPage Schema
How do I add a warning prompt before running destructive bash commands like rm -rf or git reset --hard?

You can add safety guardrails by enabling careful mode, which implements pre-run hooks that detect risky patterns in bash commands and prompts you for confirmation before execution. It catches commands like rm -rf, git reset --hard, kubectl delete, and DROP TABLE.

Can I override the safety warning for specific destructive commands when I know they are safe?

Yes, careful mode provides a per-command override option that allows you to proceed with execution when appropriate, along with a safe-exception list for trusted tasks you want to run without prompts.

Does this destructive command warning system work for kubectl delete and database operations like DROP TABLE?

Yes, the safety guardrails detect risky patterns across multiple tooling environments, warning you before executing destructive commands like kubectl delete and DROP TABLE in your bash workflows.

What is the best way to prevent accidental data loss when debugging in production environments?

The best way is to implement pre-run hooks that warn before executing destructive commands. This protects systems by reducing the risk of data loss on prod or in shared environments during collaborative debugging.

Do I need Python3 installed to use bash command guardrails?

Yes, you need Python3 installed as a dependency to run the careful safety guardrails, which detect risky command patterns and prompt for confirmation before execution in your bash workflows.