The run's priciest five minutes - 2 times a typical window. Every model call is metered, so a run's spend can be read window by window against its own baseline. For five minutes, 31 model requests from 7 agents landed at once - 2 times the run's typical five-minute spend. The burst bought something: the code-review agent finished its job.

Where this happened.

Source events.

Detected mechanically by the cost_spike detector from 2 telemetry event(s):

  • a29c64b3-c2e6-4757-8138-ff66038a6ee2
  • 59d9dbfa-20ea-40a7-a563-5f35b7141693