Everything I could decide myself is decided. What's left needs
you. Each one carries the recommendation of the session that raised it — if you
agree, pick it and press Done. A model's opinion, where there is one, appears on the card
as a receipt line; cards with no receipt were never put to a model. "I'm not sure" is a real answer and becomes work for me.
You don't need to open the session — pressing Done sends your answer back and it picks up.
1github-reviewer-down on propflowai#8278 (16:57 UTC): claude-code-review died on both picked tokens with 'You've hit your session limit · resets 7:10pm (UTC)'; sibling runs (p8-pickvendor-part1, p8-roster-p1) died on 'You've hit your weekly limit · resets 12am (UTC)'. Live probe 17:13 UTC of all 10 setup tokens: 7 at 7d=1.0 (rejected), propflow.agent + propflow.clara at 5h=1.0 (rejected), only propflow.user answering 200 (5h=0.25, 7d=0.33) with ~8 concurrent review runs drawing on it. Rotation re-run by hand at 17:15 UTC (exit 0, pool re-ranked propflow.user first). Prior block b88271588 (12d) was resolved with a question, not a capacity decision. Decision: add a reviewer account now (purchase + ccswitch capture on the mini) or ride out the resets.
In plain terms. Should we add another Claude account for the code reviewer today, or wait for the caps to reset tonight? Right now 9 of our 10 reviewer accounts are out of budget (7 spent their weekly allowance, 2 spent their 5-hour window) and one account is carrying every review. If it runs out too, no pull request can get a review until 7:10pm UTC (one account comes back) or midnight UTC (seven come back). Adding an account costs money and someone has to log it in on the mini; waiting means reviews may stay dark for a few hours on a busy Sunday.
The session raising this recommends: Wait for tonight's resets Seven accounts come back at midnight UTC and one at 7:10pm UTC; a new account bought now still needs an interactive login on the mini to be usable, so it would barely beat the reset. If this repeats next Sunday the roster is undersized and a seat is worth buying. Tried first, unsuccessfully: Read the failed run logs (session-limit + weekly-limit results on both primary and fallback), probed all 10 reviewer tokens live for 5h/7d headroom, re-ran the 2h token rotation early so the one live account ranks first, confirmed attempt three on the current head is already running (did not re-fire), searched open blocks for an existing capacity decision (b88271588 is resolved, not adoptable).