careful

Intercept destructive Bash commands and require user confirmation before execution.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/bananaduck1/storylab --skill careful-bananaduck1
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/bananaduck1/storylab/tree/main/.claude/skills/gstack/careful
Command: npx skills add https://github.com/bananaduck1/storylab --skill careful-bananaduck1

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python3.

What problem does it solve?

This guardrail skill intercepts destructive shell commands and prompts for confirmation before execution, reducing the risk of data loss in shared or production environments.

Core Features & Use Cases

  • Pre-execution warnings for dangerous operations (rm -rf, git reset --hard, kubectl delete, etc.)
  • Per-command override capability to proceed after confirmation
  • Built-in PreToolUse hook for Bash to intercept inputs in real-time
  • Suitable for development, staging, and production contexts where irreversible actions must be guarded

Quick Start

Enable careful mode and run a command you would normally execute to see the warning 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 warning prompt before executing destructive bash commands?

To add a warning before destructive bash commands, use a PreToolUse hook to intercept shell inputs and require explicit user confirmation before operations like rm -rf or git reset --hard proceed.

Can I override the safety check on a per-command basis for specific bash operations?

Yes, you can override the safety check on a per-command basis, allowing you to bypass the guardrail and proceed with executing destructive operations after providing explicit confirmation.

How do I prevent accidental data loss from destructive commands in production environments?

Prevent accidental data loss in production by enforcing pre-execution warnings that intercept destructive shell commands, requiring confirmation before operations can execute on sensitive servers or shared workspaces.

Does this destructive command guard require Python to run?

Yes, this destructive command guard requires Python3 installed in your environment to power the PreToolUse hook that intercepts and evaluates bash inputs in real-time.

What types of destructive operations does a pre-execution bash hook intercept?

A pre-execution bash hook intercepts destructive operations such as file deletions, hard git resets, and resource deletions, prompting for user confirmation before allowing the command to execute.

When should I use a prehook to guard shell commands instead of manual review?

Use a prehook to guard shell commands in development, staging, or production contexts where irreversible actions occur, ensuring real-time interception that manual review processes can easily miss.