careful

Detect destructive Bash commands and warn before execution.

3|3|Updated Mar 26, 2026
One-click install
npx skills add https://github.com/tsensei/ReelMistri --skill careful-tsensei
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/tsensei/ReelMistri/tree/main/.claude/skills/gstack/careful
Command: npx skills add https://github.com/tsensei/ReelMistri --skill careful-tsensei

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Safety guardrails for destructive commands ensure users are warned before executing dangerous operations, preventing accidental data loss in development, staging, and production environments.

Core Features & Use Cases

  • Pre-use guard checks: A pre-tool hook evaluates Bash commands for destructive patterns and warns or blocks them before execution.
  • Override capability: Users can opt to proceed after a warning when they consciously intend to perform a risky action.
  • Wide applicability: Useful in CI pipelines, interactive shells, and remote sessions to minimize accidental destructive actions.

Quick Start

Enable careful mode and run a command; if a dangerous command is detected, you will receive a warning and may choose to proceed.

Frequently Asked Questions about careful

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

FAQPage Schema
How do I prevent accidental data loss from destructive Bash commands in production?

You can prevent accidental data loss by using safety guardrails that detect destructive Bash commands and issue warnings before execution. A pre-use hook evaluates shell commands for risky patterns, allowing you to block or override dangerous actions in staging and production environments.

Can I override a safety warning if I intentionally need to run a risky shell command?

Yes, you can override safety warnings when you consciously intend to execute a risky shell command. The guardrail system prompts you before proceeding, ensuring the destructive action is deliberate while maintaining a safety net against accidental execution.

Do destructive command guardrails work in CI pipelines and interactive shells?

Destructive command guardrails work across CI pipelines, interactive shells, and remote sessions. The pre-use hook evaluates Bash commands in any shell workflow, minimizing accidental destructive actions wherever risky commands might destabilize environments or delete data.

What is a pre-use hook for evaluating destructive commands?

A pre-use hook for destructive commands is a mechanism that intercepts Bash commands before execution to scan for risky patterns. It applies safety guards by warning or blocking the operation, requiring a YAML frontmatter with name and description to function.

How to set up safety guards for destructive commands in automated workflows?

To set up safety guards for destructive commands, enable careful mode and execute your shell commands. The pre-use hook automatically evaluates the commands for destructive patterns, issuing warnings that you can choose to override if the risky action is intentional.

What are the limitations of using a pre-use hook to block destructive Bash commands?

The limitation of using a pre-use hook to block destructive Bash commands is that it relies on pattern detection, which requires a YAML frontmatter for configuration. It cannot prevent damage if a user consciously overrides the warning and authorizes a risky action.