A customer is actively asking for support. What should the assistant do next?
The workflow
Eleven readings of the conversation at once: what the customer wants and how they want it resolved, how frustrated they sound, how urgent it is, and flags such as unauthorized activity, a legal threat, or a request for a person.
Four follow-ups, plus a tie-break when two intents are close. Each runs only when the first reading calls for it, with the records it needs attached: the pending proposal, the ledger and card status, the refund or fee on file, the subscription.
The assistant's own messages are compared with the record: did it tell the customer a refund had gone through, or that the card was frozen, when neither had happened?
Nine sections run top to bottom: safety, the assistant's claims, a pending proposal, handing off, closing, then the issue itself. The first rule that applies in a section emits its actions; when none applies, the next section runs.
A handful of cases: for each model, one where it alone differs from the other two; one where all three miss the reference; one where all three agree. Each model's path is drawn on the chart in its own colour and line style. Where paths share a stretch, the lines run side by side. What no model reached is faded. Underlined questions open below with every model's answer and the state it read.