Start here
The free tool below runs entirely in this browser — no account, nothing stored or transmitted. Redact anything you would not want kept before you type it.
Open the companion tool →Why the scope is the whole decision
An AI agent's permission prompt is the moment the scope is set: what it can spend, up to what amount, for how long, and whether it can act without asking again. After you tap allow, the agent can act inside that scope without you.
The reading is the same discipline as any authority you grant: name the limit, the duration, and the way to revoke it. If the prompt is vague about any of those, that is the reading — do not authorize it.
The protocol
- Enter or describe the permission prompt — the amount, duration, and what it may do.
- Read the scope basis: spend limit, duration, and revocation each get a line.
- If any of the three is vague or missing, do not authorize yet.
- Use a card with a low limit, and set the shortest duration that still works.
What the tool does
Enter an AI agent's permission prompt and get a scope reading with spend, duration, and revocation broken out. It runs in this browser; nothing is sent or stored.
Limit first: This is an authorization-reading aid, not financial or legal advice. It cannot see the platform's actual terms. Verify the current permission scope in the platform itself.
Open the companion tool →One human next step
Authorize only the minimum scope, on a low-limit card, and confirm the revocation route in the platform's settings before the first purchase.
Research log and safety checks
AI-assist path: not used for a verdict. This artifact is an inspectable scope reading; a remote model would add an opaque interpretation where the reader needs to see the exact scope wording. §6.10: not a frontier-pace topic.
§0.5 protection result: category is financial-agent-exposure; crisis-adjacent: no; fear/urgency/scarcity toolkit used: no. The protective function is free and the paid feature is convenience only.