careful

Monitors commands for destructive operations and requires explicit user confirmation before execution.

Updated Jun 21, 2026
One-click install
npx skills add https://github.com/apexradius/agentic-os --skill careful-apexradius
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/apexradius/agentic-os/tree/main/skills/careful
Command: npx skills add https://github.com/apexradius/agentic-os --skill careful-apexradius

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill acts as a safety net for production environments, alerting users to potential catastrophic commands and ensuring that destructive actions are confirmed before execution.

Core Features & Use Cases

  • Prevents Accidental Destructive Actions: Alerts users about dangerous commands like 'rm -rf', 'DROP TABLE', 'git reset --hard', etc., before they are executed.
  • Explicit Confirmation: Requires users to confirm destructive actions, reducing the risk of accidental data loss.
  • Use Case: Use 'careful' in shared environments, production servers, or for personal workflows to reduce the risk of unintended data deletion or system disruption.

Quick Start

Set up 'careful' mode by activating it when necessary or by configuring your environment to automatically enable it.

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 commands like rm -rf or DROP TABLE in production?

To prevent accidental data loss from destructive commands, you should use a safety guardrail mechanism that monitors execution and requires explicit confirmation before proceeding with operations like 'rm -rf' or 'DROP TABLE'.

Can I require explicit confirmation before executing destructive shell commands?

Yes, requiring explicit confirmation before executing destructive shell commands is achievable by activating a safety mode that alerts you to catastrophic operations and pauses execution until you manually confirm the action.

Does command validation work for shared environments to prevent system corruption?

Command validation works effectively for shared environments by monitoring execution and requiring confirmation for destructive actions, thereby minimizing unintended data deletion and system corruption across shared infrastructure and production servers.

What is the best way to add destruction guardrails for git reset --hard?

The best way to add destruction guardrails for 'git reset --hard' is to activate a command safety mode that monitors for potentially destructive operations and intercepts them, requiring explicit user confirmation before the command proceeds.

How do I set up confirmative execution for dangerous commands?

You can set up confirmative execution for dangerous commands by activating a safety mode or configuring your environment to automatically enable it, which monitors operations and requires explicit confirmation before executing destructive actions.

When should I not use command safety monitoring?

You should not use command safety monitoring in fully automated, non-interactive pipelines where execution must proceed without waiting for explicit user confirmation, as the guardrail requires manual input to authorize destructive actions.