3 agents of one type fired into the same trap - each replaced in minutes. The orchestrator can fire an agent mid-run and hire a same-type replacement without stopping the job. The code-review agent ran out of room in its context window and was removed. A replacement was hired at once - and 3 same-type agents went down the same way inside 3 minutes.

Where this happened.

Source events.

Detected mechanically by the restaff_chain detector from 6 telemetry event(s):

  • 03f6e492-56f7-4f0e-b6dd-9f9be03e6f41
  • 88568ca6-8d44-4958-985f-b92bcb3b18bb
  • f76e2b76-f3b0-414f-90b5-4213527c1a87
  • c42279b9-3b1a-4e70-be90-c253540c3c1d
  • 63fc677c-c03e-4cd6-891d-a65655f2063a
  • 732e6ca3-74e7-4398-8aa7-394861e5f8d5