careful

Detect and block destructive bash commands before execution.

Updated Mar 25, 2026
One-click install
npx skills add https://github.com/fghaffar/faystack --skill careful-fghaffar
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/fghaffar/faystack/tree/main/careful
Command: npx skills add https://github.com/fghaffar/faystack --skill careful-fghaffar

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents accidental execution of high‑risk operations such as recursive deletes, database drops, force pushes, and cluster deletions, protecting production and shared environments from irreversible damage.

Core Features & Use Cases

  • Pattern Detection: Scans Bash commands for dangerous keywords and flag combinations.
  • Safe Exceptions: Allows harmless deletions of common build artefacts without warnings.
  • User Prompt: Provides a clear warning and lets the user decide whether to proceed.
  • Use Cases: Ideal for developers working on live servers, CI pipelines, or any context where a stray destructive command could cause data loss.

Quick Start

Ask the careful skill to run a command, and it will warn you if the command is destructive.

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 or git force-push from running?

To prevent destructive bash commands like rm -rf or git force-push, you need a safety mechanism that scans command JSON for dangerous patterns and prompts for user permission before execution. This skill intercepts high-risk operations and returns clear warnings.

What destructive command patterns does a bash safety scanner need to detect?

A bash safety scanner should detect destructive command patterns such as rm -rf, DROP TABLE, git force-push, and cluster deletions. Recognizing these dangerous keywords and flag combinations prevents irreversible damage to production and shared environments.

Can I block destructive commands in shared environments and CI pipelines without blocking safe deletions?

Yes, you can block destructive commands in shared environments and CI pipelines while allowing safe exceptions. The scanner identifies harmless deletions of common build artifacts and lets them run without triggering user warnings.

How do I get a warning before executing dangerous kubernetes or docker commands?

Getting a warning before executing dangerous kubernetes or docker commands involves analyzing the command against a destructive pattern database. The scanner detects high-risk flags and prompts you to decide whether to proceed before execution.

Does a command blocking tool require complex dependencies to intercept bash commands?

A command blocking tool does not require complex dependencies to intercept bash commands. This skill requires only Bash tool access to read command JSON, evaluate patterns, and return permission decisions without additional environment setup.