careful

Detect destructive commands like 'rm -rf' and prompt for confirmation.

Updated Mar 27, 2026
One-click install
npx skills add https://github.com/BaGyun0107/harness-starterkit --skill careful-bagyun0107
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/BaGyun0107/harness-starterkit/tree/main/.agents/skills/gstack-careful
Command: npx skills add https://github.com/BaGyun0107/harness-starterkit --skill careful-bagyun0107

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill prevents accidental or malicious execution of destructive commands by detecting harmful patterns before commands run.

Core Features & Use Cases

  • Destructive Command Detection: Identifies risky commands like 'rm -rf', 'DROP TABLE', or 'git push --force' before execution.
  • User Warnings & Overrides: Prompts for confirmation when dangerous commands are detected, reducing unintended data loss or system damage.
  • Use Case: When working on production servers or shared environments, users can run commands confidently knowing that safeguards are active to prevent accidental harm.

Quick Start

Use the careful skill to review a bash command, warn for destructive patterns, and ask for user confirmation.

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 deletion from destructive bash commands?

To prevent accidental data deletion from destructive bash commands, you can use an automated warning system that detects risky patterns like 'rm -rf' and prompts for confirmation before execution. This safeguards against unintended system damage.

What is the best way to add a safety warning before running 'DROP TABLE' or 'git reset --hard'?

The best way to add a safety warning for commands like 'DROP TABLE' or 'git reset --hard' is to implement pattern matching checks that intercept these commands and require user confirmation to prevent accidental database or repository damage.

Can I intercept and block destructive commands in a DevOps workflow?

Yes, you can intercept and block destructive commands in a DevOps workflow by applying automated pattern matching to identify harmful commands, prompting for user overrides to ensure safety during execution on shared environments.

Does pattern matching for command safety work for both system administration and database management?

Pattern matching for command safety works effectively for both system administration and database management by detecting risky command patterns specific to each domain, such as 'rm -rf' for systems and 'DROP TABLE' for databases, preventing accidental harm.

How do I set up an override prompt for risky commands on production servers?

To set up an override prompt for risky commands on production servers, configure a safety skill that reviews commands, warns for destructive patterns, and asks for explicit user confirmation before allowing execution to proceed.

What are the limitations of using pattern matching for destructive command detection?

The limitation of using pattern matching for destructive command detection is that it primarily relies on identifying known risky patterns like 'rm -rf', meaning novel or obfuscated destructive commands might bypass the matching rules if they do not fit expected shapes.