Scan unlabeled traffic for suspected failures
Judges (criteria) API
Scan unlabeled traffic for suspected failures
Judges a bounded batch of recent UNGRADED logged traffic with this judge; FAILs become pending suspects for one-click confirm/dismiss in the dashboard’s Review queue. Confirming creates a real fail grade the calibration set grows by; dismissing means it never resurfaces — the judge suggests, the human decides. Candidates are the most recent logged traffic minus already-graded and already-scanned, scoped to the judge’s population (tag and segment). Trace-unit judges are rejected — grade whole runs in the Review queue instead. Spends the wallet like any judging (owner/admin key).
POST
Scan unlabeled traffic for suspected failures