careful

Warn and require confirmation before destructive shell commands execute.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Destructive commands pose a risk to production systems and shared environments; this skill adds a guardrail that warns and can block or require confirmation before dangerous actions.

Core Features & Use Cases

  • Real-time warning for destructive patterns like rm -rf, git reset --hard, kubectl delete.
  • Optional override to proceed after a warning.
  • Applies in development, staging, or production debugging contexts where destructive commands risk data loss.

Quick Start

Enable careful mode in your shell workflow and run a command to trigger the guardrail warnings before any destructive operation.

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?

To add a warning prompt before destructive bash commands, implement pattern-based checks that detect rm -rf and prompt for confirmation before execution. This guardrail applies in interactive shells and automated scripts to prevent accidental data loss.

Can I safely override a guardrail warning if I need to force a git reset --hard?

Yes, you can safely override a guardrail warning to proceed with a git reset --hard. The override mechanism allows you to bypass the confirmation prompt after acknowledging the risk of the destructive command.

Does this command safety check work with kubectl delete operations in automated scripts?

Yes, this command safety check works with kubectl delete operations in automated scripts. It provides a JSON-based permission response that integrates with automated frameworks to warn and require confirmation for dangerous actions.

How do I configure a safe-exceptions list to whitelist specific destructive commands?

You can configure a safe-exceptions list to whitelist specific destructive commands by defining approved patterns. Commands matching these exceptions bypass the warning and confirmation prompts, allowing execution without triggering the guardrail.

When do I need a destructive command guardrail in production debugging contexts?

You need a destructive command guardrail in production debugging contexts when actions like rm -rf, git reset --hard, or kubectl delete risk irreversible data loss in shared environments. The guardrail ensures you confirm before executing dangerous operations.