Compare an agent's quality over two windows
Compute the agent’s continuous quality scorecard: rolls the recent ‘current’ window (default 7 days) against the equally-sized ‘baseline’ window that precedes it across accuracy (rubric pass rate), sentiment, p50/p95 latency, thumbs-down rate, and error rate, and flags any regression above noise thresholds. Call it the moment after a prompt or model change to catch silent degradation the volume/cost analytics miss; it is also the signal the canary-rollout auto-advance / auto-rollback gate reads (optional canary_stages / canary_stage / canary_rollback_on_warn query params return a staged rollout decision alongside the scorecard). Requires the agents:read scope; read-only.
Authorizations
Dashboard JWT token from Clerk
Headers
Sandbox opt-in for Clerk-session-authenticated requests. Set to true to route the call through the test-mode pipeline: no real provider delivery, no credits deducted, response meta.test_mode: true. Ignored for live API keys (dv_live_sk_*) — server-to-server clients must use a test-prefixed key (dv_test_sk_*) to exercise sandbox. Test-prefixed keys unconditionally enable sandbox regardless of this header.
true, false Path Parameters
Response
Successful response.