careful

Warn before executing dangerous Bash commands with per-command overrides.

Updated Mar 21, 2026
One-click install
npx skills add https://github.com/TimHL5/carousel --skill careful-timhl5
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/TimHL5/carousel/tree/main/.claude/skills/gstack/careful
Command: npx skills add https://github.com/TimHL5/carousel --skill careful-timhl5

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents accidental destructive actions by warning before dangerous commands are executed and requiring explicit override when needed.

Core Features & Use Cases

  • Real-time warnings for dangerous commands (rm -rf, git reset --hard, kubectl delete, etc.).
  • Per-command override options to allow trusted operations.
  • Works with Bash-based workflows across local machines, servers, and CI environments.
  • Useful for prod maintenance, debugging in shared environments, or any scenario requiring careful command execution.

Quick Start

Activate careful mode and attempt a destructive command to see the warning and optionally override

Frequently Asked Questions about careful

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

FAQPage Schema
How do I add safety guardrails to prevent destructive bash commands?

To add safety guardrails for destructive bash commands, you need a warning system that intercepts dangerous operations like rm -rf before execution. This Skill implements a PreToolUse hook that detects risky commands and requires explicit override to proceed.

Can I override the warning when executing destructive commands in prod mode?

Yes, you can override the warning for destructive commands in prod mode using per-command override options. This allows trusted operations to proceed after the initial guardrail warning is triggered, ensuring intentional execution of necessary destructive actions.

Does this safety hook work with kubectl delete and git reset --hard?

Yes, this safety hook works with kubectl delete and git reset --hard. It applies to Bash-based workflows and covers a wide range of destructive operations across local machines, servers, and CI pipelines by issuing real-time warnings.

What is the best way to prevent accidental destructive actions in CI pipelines?

The best way to prevent accidental destructive actions in CI pipelines is using a PreToolUse hook with YAML frontmatter configuration. It monitors Bash-based workflows across shared environments and provides real-time warnings before dangerous commands execute.

How do I configure the warning behavior for specific bash commands?

You configure warning behavior for specific bash commands using YAML frontmatter. This configuration supports per-command override behavior, allowing you to define exactly which destructive operations require warnings and how to handle explicit overrides.

When should I use a bash command guardrail in shared environments?

You should use a bash command guardrail during prod maintenance, debugging in shared environments, or any scenario requiring careful command execution. It prevents accidental destructive actions by warning before dangerous commands are executed across machines and servers.