Why, and what was tried
The session raising this recommends: Yes — point it at the real work
The fix moved to Trinity's PR on Fede's call, so the check naming my own branch can never pass — that branch never gets created. Pointing the finish line at the three real things (their fix merged, our write-up merged, the permissions change merged) covers everything the job was actually asked for. Only you can shrink the list; I can only add to it.
Tried first, unsuccessfully: fable-decide ran twice on exactly this. Receipt f7ddfc33b came back RESOLVED recommending 'Escalate now: raise a block asking the human to drive --replace the criterion' — the model's own ruling is that this needs you, and I am executing it, not overriding it. Free-text only because --fable-receipt refuses a RESOLVED receipt by design, and the tool names this as the channel. It independently verified the same two facts I did: _eval_run_green (agent-smith activities/operator_task.py:1541) fails closed on zero runs, and only drive --replace may shrink the list. It added one I had missed: a dead item beside passing ones recreates the recorded count-mismatch hazard where the gate reads 2/3 while the finisher reads 3/3. Suggested --replace wording: 'Removing then re-adding a teammate in the same company leaves exactly one Person with the role active and phone intact, proven by Trinity propflowai#20497 merged to main; plus the written design-fit check published (propflow-docs#365 merged). No prod data repair performed.'
0d5d38b8 is parked on this.