skills.rest
Dragoon0x
Dragoon0xCOMMUNITY
Explore all skills by this author →
Website𝕏@0xDragoonLabGitHub

careful

Generates a .dragoon/CAREFUL.md checklist of destructive commands requiring explicit user approval.

4|Updated Apr 11, 2026
Search Tags:#onboarding#guardrails#risk prevention#destructive commands#codebase safety#ai agent discipline#command checklist
One-click install
npx skills add https://github.com/Dragoon0x/dragoon-skills --skill careful-dragoon0x
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/Dragoon0x/dragoon-skills/tree/main/skills/careful
Command: npx skills add https://github.com/Dragoon0x/dragoon-skills --skill careful-dragoon0x

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

AI agents and new contributors often accidentally execute irreversible destructive commands like rm -rf, force pushes, or database drops, leading to catastrophic data loss with no pre-execution warning system in place.

Core Features & Use Cases

  • Destructive Command Checklist Generation: Creates a .dragoon/CAREFUL.md file listing commands that require explicit user confirmation before execution, and commands that are never allowed without direct user approval.
  • Onboarding & Incident Prevention: Ideal for onboarding new AI agents to a codebase, codifying rules after a destructive accident, and establishing explicit team guardrails for high-risk operations.

Quick Start

Use the careful skill to generate a destructive command guardrail checklist for your project.

Frequently Asked Questions about careful

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

FAQPage Schema
How do I prevent AI agents from running destructive commands like rm -rf or force pushes?▼

You can prevent accidental data loss by generating a .dragoon/CAREFUL.md checklist that lists destructive commands requiring explicit user confirmation before AI agents or new contributors execute them.

What is a destructive command guardrail checklist for codebase onboarding?▼

A destructive command guardrail checklist is a .dragoon/CAREFUL.md file codifying rules for high-risk operations like file deletion, force pushes, and database drops to prevent accidental irreversible data loss during onboarding.

How do I create a command safety checklist after a destructive codebase accident?▼

You create a command safety checklist by generating a .dragoon/CAREFUL.md file that codifies post-accident rules, listing destructive commands that are never allowed without direct user approval to prevent future incidents.

Does a destructive command checklist actually block runtime command execution?▼

No, the generated checklist provides no runtime command blocking functionality; it establishes a read-before-act discipline file listing destructive commands that require explicit user confirmation before AI agents execute them.

Can I use a command guardrail checklist for new contributor onboarding?▼

Yes, the skill is ideal for onboarding new contributors to a codebase by generating a checklist of destructive commands requiring explicit user confirmation, establishing explicit guardrails for high-risk operations like file deletion and database drops.

Related Skills

Explore More Software Engineering Skills→
Dragoon0xDragoon0x
4

benchmark

Capture timestamped JSON snapshots of codebase size and design critique scores.

Community
Intermediate
akunikkolaakunikkola
97

tekoaly-velvoitteet

Determine EU AI Act obligations and compile technical documentation for high-risk systems.

Community
Advanced
oioio-spaceoioio-space

repo-janitor

Remove build and scan artifacts before commits via pre-commit hooks.

Community
Intermediate
GenEducationGenEducation

guard

Warn on destructive shell commands and restrict file edits to an approved directory.

Official
Intermediate