Arize Phoenix TS
    Preparing search index...

    One aggregate acceptance rule, evaluated once after every test in the suite has run. Each criterion aggregates a single annotation's scores with one AcceptanceMetric and fails the suite when the result misses its bar.

    Scoring notes shared by every metric:

    • Boolean scores count as 1 (true) / 0 (false).
    • If a run logs the same annotation more than once, the last one counts.
    • Skipped tests are excluded; dry-run tests are included (they still run).
    • A criterion whose annotation was never logged on any run fails (rather than passing vacuously) — see AcceptanceResultFields.failureReason.