Cap the intent. Keep the provider key.

Isolate the intent traffic, not the provider account.

1. 01
   Issue a disposable Till key for the intent job

   Set the request ceiling from expected utterances and retries, then add optional token, estimated-spend, expiry, and IP bounds when the classifier needs a tighter envelope.

2. 02
   The provider credential stays behind Till

   The intent job uses a Till key, not an unconstrained provider secret in the worker process. When the job is retired, revoke or let expiry stop that key without rotating every connection.

3. 03
   This is not an NLU engine or intent catalog

   Till does not classify utterances, train intent models, or replace your router. It is a hosted control layer in front of 12 AI providers: a required request ceiling and optional token, spend, expiry, and IP controls.

4. 04
   Pro: $59/month, 25 scoped keys, 25,000 monthly activations

   Those are plan limits. Public signup is closed; request Pro beta access. Provider inference is billed by the provider. Till charges a platform subscription.