Profile My skills Saved Collections Edit profile Submit a skill

Careful

Write a checklist of destructive commands the user wants to think twice about (rm -rf, force push, hard reset, db drop, etc.) to .dragoon/CAREFUL.md. dragoon does NOT block commands at runtime - this is a discipline tool, a list the agent reads before risky operations. Use to set explicit guardrails for an AI agent or new contributor. Run as `node ~/.claude/skills/dragoon/skills/careful/scripts/careful.js`.

dragoon0x 188106d 2 files · 3.8 KB Updated
AI & ML

File contents

/careful

Discipline-tool checklist for destructive commands.

When to use

  • Setting up a new project or onboarding an agent
  • After a destructive accident, codify the rule
  • The user says "set up guardrails", "what should the agent not do"

What it writes

.dragoon/CAREFUL.md with two sections:

  • destructive commands to confirm before running (rm -rf, force push, etc.)
  • destructive commands the agent should NEVER run without explicit user prompt

This is not a runtime block. It's a read-before-act discipline aid.

Run it

dragoon careful --apply
dragoon careful --show

Dragoon0x/dragoon-skills/tree/main/skills/careful commit 188106d4c3

Browse category

Frequently asked questions

npx skillmds@latest add dragoon0x/careful