Day-bucketed eval-run activity
One row per UTC day over the requested window, for the activity calendar — so a 12-month chart costs ~365 small rows instead of every raw eval run. Every day is present, including zero days. failed uses the same definition as GET /v1/eval-runs?outcome=failed (a completed run with at least one failing score), so clicking a cell and drilling into the list agrees with the chart. Server-side cached for 60s; the window is capped at 400 days.
Authorizations
Query Parameters
Inclusive lower bound; defaults to 365 days ago.
Upper bound; defaults to now. A date-only value (midnight) includes that whole day, matching the list endpoint.
live, simulation, test Response
Successful Response
Day-bucketed eval-run counts over a window, for the activity calendar.
Days are UTC. Every day in [period_start, period_end) is present,
including zero days, so a calendar grid needs no client-side gap filling.