The run's priciest five minutes - 5.4 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, 107 model requests from 13 agents landed at once - 5.4 times the run's typical five-minute spend. The burst bought something: the sprint-plan agent finished its job.

Where this happened.

Source events.

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

  • 67180d78-ad41-40f9-acbd-923a937f71f1
  • b3181394-c97a-487e-9e5a-f7ce091a77ce