careful

Detect dangerous Bash commands and warn before execution.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/effiraid/reflix-nextjs --skill careful-effiraid
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/effiraid/reflix-nextjs/tree/main/.claude/skills/gstack/careful
Command: npx skills add https://github.com/effiraid/reflix-nextjs --skill careful-effiraid

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Protect systems by detecting dangerous, destructive commands in Bash and warn before execution. Users can override warnings if they accept the risk.

Core Features & Use Cases

  • PreToolUse hook for Bash that scans commands for risky patterns such as rm -rf, git reset --hard, git push --force, kubectl delete, and docker system prune.
  • User-driven overrides allow proceeding after a warning, or canceling to preserve safety.
  • Suitable for development, staging, and production workflows where destructive actions must be guarded.

Quick Start

Enable careful mode before running risky commands to receive warnings and opt to proceed or cancel.

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 Bash commands like rm -rf in production?

To prevent destructive Bash commands, you need a PreToolUse hook that scans input for risky patterns like rm -rf and warns before execution. This skill intercepts commands across local, staging, and production environments, allowing you to cancel or accept the risk.

Can I get a warning before running git reset --hard or git push --force?

Yes, you can get a warning before running git reset --hard or git push --force by using a Bash command scanner. It detects these specific risky patterns and prompts you with an override or cancel option to preserve system safety.

Does this safety guardrail work with Kubernetes and Docker cleanup commands?

Yes, this safety guardrail works with Kubernetes and Docker cleanup commands by detecting operations like kubectl delete and docker system prune. It scans these Bash-driven inputs and allows you to override the warning or cancel the action entirely.

How do I override a destructive command warning if I accept the risks?

You can override a destructive command warning by choosing to proceed when prompted. The user-driven override feature allows you to accept the risk and execute the Bash command, or cancel it to maintain production safety guardrails.

Are there limitations when using Bash hooks to block destructive commands?

A limitation of using Bash hooks to block destructive commands is that they rely on pattern matching for known risks like git push --force. Users must manually evaluate and accept risks during overrides, and configure optional exemptions for specific workflows.