Chronomata persistent agents
Both conditions answered all six known-fixture tasks correctly after an explicit recovery amendment. Persistent observation memory used three calibrations versus six, but required 37 inference dispatches versus 32. Canceled-call uncertainty and every earlier outcome remain preserved.
Chronomata persistent agents
A completed 4B delivery calibration met the 30-second deadline on all three requests, but produced no valid investigation decision. This is a failed action-delivery result, not a memory or learning demonstration.
Developmental computation
Original assessment and bounded engineering witness; both comparative protocols were unexecuted.
Developmental computation
Two exploratory seeds and three conventional arms. Ordinary growth retained outputs/state exactly; the added branch lowered squared loss but hurt query-sign accuracy.