careful

Warn before destructive shell commands require explicit user confirmation.

1|Updated Mar 31, 2026
One-click install
npx skills add https://github.com/LaPaGaYo/nexus --skill careful-lapagayo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/LaPaGaYo/nexus/tree/main/skills/safety/careful
Command: npx skills add https://github.com/LaPaGaYo/nexus --skill careful-lapagayo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Destructive operations in shells and production environments happen accidentally; this feature adds pre-execution guardrails to warn users and prevent unintentional data loss.

Core Features & Use Cases

  • Warning prompts for dangerous commands (rm -rf, git reset --hard, kubectl delete, DROP TABLE, etc.).
  • User-confirmation overrides to proceed when explicitly authorized.
  • Session-scoped safety mode that can be toggled per session for debugging or prod work.

Quick Start

Activate careful mode in your current shell session and try a destructive command to see the warning prompt.

Frequently Asked Questions about careful

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

FAQPage Schema
How do I prevent accidental destructive shell commands like rm -rf or git reset --hard from executing?

To prevent accidental destructive shell commands, you need pre-execution guardrails that detect risky patterns like rm -rf or git reset --hard and prompt for explicit user confirmation before execution, preventing unintentional data loss.

What destructive command patterns can a shell guardrail detect?

A shell guardrail can detect destructive command patterns including rm -rf, DROP TABLE, git reset --hard, kubectl delete, and docker rm -f. It implements a pre-command hook to intercept these risky operations and prompt for confirmation in production or shared environments.

How do I add a warning prompt before running kubectl delete or docker rm -f in production?

To add a warning prompt before running kubectl delete or docker rm -f, activate a session-scoped safety mode that implements a pre-command hook. This detects risky patterns and requires explicit user confirmation to proceed when authorized, safeguarding production environments.

Can I override the shell safety warning when I intentionally need to run a destructive command?

Yes, you can override the shell safety warning using user-confirmation overrides to proceed when explicitly authorized. The session-scoped safety mode can be toggled per session, allowing you to bypass the prompt for debugging or production work when necessary.

Does a pre-command hook work with Bash workflows in shared environments?

Yes, a pre-command hook works with Bash and shell workflows in shared environments. It is specifically applicable to production or shared environments where destructive operations may jeopardize data, adding guardrails to warn users before execution.