careful

Warn before executing destructive shell commands like rm -rf and kubectl delete.

1|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/cheng80/rummipoker --skill careful-cheng80
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/cheng80/rummipoker/tree/main/.agents/skills/gstack-careful
Command: npx skills add https://github.com/cheng80/rummipoker --skill careful-cheng80

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Safeguards prevent destructive shell operations by warning users before executing commands like rm -rf, DROP TABLE, git reset --hard, or kubectl delete, reducing accidental data loss in shared or prod environments.

Core Features & Use Cases

  • Provides per-command safety checks with override options for urgent prod debugging.
  • Detects common destructive patterns in Bash, Git, and database commands and prompts for confirmation before continuing.
  • Use Case: A developer runs a script that suddenly includes rm -rf /, and the safety guard stops it, asking for explicit confirmation.

Quick Start

Enable safety mode for destructive commands in your shell sessions and confirm any warnings before execution.

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 shell commands like rm -rf in shared environments?

Destructive shell command guardrails prevent accidental data loss by intercepting risky commands like rm -rf or git reset --hard, prompting for explicit user confirmation before execution proceeds in shared or production environments.

How does pattern-based detection work for intercepting dangerous Bash, Git, and database operations?

Pattern-based detection for dangerous shell operations works by matching command strings against known destructive patterns in Bash, Git, and database queries, using safe-execution hooks to pause execution and issue user-configurable warnings.

Can I override shell safety warnings for urgent production debugging when needed?

Shell safety warnings support controlled overrides, allowing developers to bypass guardrails and execute intercepted destructive commands when urgent production debugging requires immediate action.

Does this shell guardrail tool protect automated scripts that suddenly include destructive commands?

Shell safety guardrails protect automated scripts by intercepting commands containing destructive patterns like rm -rf /, stopping execution, and requiring explicit confirmation before the script can proceed.

What are the limitations of using user-configurable warnings for command safety checks?

User-configurable warnings for command safety checks rely on pattern-based detection, meaning highly novel or obfuscated destructive commands outside configured patterns may bypass the guardrails and execute without interception.