Typed events: intent, entities, urgency

Every inbound is parsed into a structured event before your agent sees the body.

Mails.ai — Typed events: intent, entities, urgency

An inbound does not arrive as text for you to classify. It arrives as an event with the classification already done: what the sender wants, what facts they named, how urgent it reads, and how much you should trust it.

The live events stream with intent, urgency, injection score and sender reputation per event

What is on a reply.received event

  • intent — what the message is trying to do, in one label.

  • entities — the structured facts pulled out of it, like a proposed time.

  • urgency — 0 to 1.

  • injection_score — 0 to 1, plus injection_categories when it is high.

  • sender_reputation — what we know about this sender.

  • extracted_text — the reply itself, quoted history removed.

The full event list

These are the event types we actually emit — subscribing to anything else would create an endpoint that can never fire, so the API rejects it:

message.sent          message.received                 draft.scheduled
message.scheduled     message.received.unauthenticated draft.sending
message.delivered     reply.received                   draft.sent
message.bounced       suppression.added                draft.failed
message.complained    webhook.test

The Events page streams them live, which is the fastest way to watch what your agent is actually receiving while you build.

Watch the event stream
Did this answer your question?
😞
😐
😁