My Alerts (beta)

Everything on this dashboard is a feed you filter while you are looking at it. The filters are good, and they are useless the moment you close the tab.

The data was never the ceiling — attention was. A contract can clear every threshold you care about at 15:15 and you will only know if you happened to be on the right page at the right minute.

My Alerts, behind the bell in the top corner, turns your definition of interesting into something that runs without you. Name a rule, stack a few conditions — premium above $500k, expiry within a day, V/OI over 2 — and it is evaluated about once a minute against the whole scanned universe.

Conditions combine with AND, never OR. Two independent triggers are two alerts. That is on purpose: a rule you can read in one line is a rule you can trust at 09:35.

The field list offers exactly what the scanner measures: premium, volume, V/OI, days to expiration, % out of the money, call or put, fill direction and bull/bear read.

Nothing else is offered. A criterion we cannot actually evaluate would produce a rule that never fires and never explains why — so those fields are absent rather than present and broken. Price action, levels and dark pool are listed as coming soon for the same reason.

Pick email, browser notification, or both. Browser asks for permission the moment you tick the box and not a second earlier — an unsolicited prompt is denied once and for all, and cannot be asked again.

Leave the ticker box empty and the rule watches everything. That is only usable because of the frequency cap: one rule can fire at most once per contract per window, 30 minutes by default. Without it, watching the whole universe during an active session would mail you continuously.

Recently triggered lists the last 50 firings with their contract and timestamp, so a rule that is too loose or too tight shows itself in minutes rather than in a week of silence.

An alert older than five minutes is dropped instead of delivered. A stale flow signal presented as fresh is worse than no signal — it would have you act on a move that is already over.

The evaluation loop is new and still being settled. Alerts can fire late, and they can fail to fire at all. Treat this as a second pair of eyes, not as the first — keep watching the pages you already watch.

That warning is not boilerplate. On an alerting tool the dangerous failure is silence: a missed notification looks exactly like a quiet market, and someone who believes they are covered when they are not is worse off than someone who knows they are watching alone. Tell us what misfires — a rule that should have fired and did not is the most useful thing you can report.