Runs & reports

Executing suites and reading scored results.

Executing

From the suite page (Execute run) or the API. The run returns an id; poll status or wait for a webhook.

Result rows

Per-case rows include execution status, QA status, score, latency, failure reason and evaluation detail (judge reasons, deterministic checks).

QA report

GET /api/v1/test-runs/{id}/report returns overall score, pass rate, counts (passed/failed/incomplete), critical and high-severity failure counts, execution/evaluation error counts and dimension breakdowns.

Report rendering in the app maps these fields directly; if a field is absent in the API response the UI shows an em dash rather than a guessed value.