generated: '2026-07-21'
method: searched
source: https://docs.startree.ai/api-reference/introduction
authentication:
  style: bearer-jwt
  detail: Bearer token (JWT) generated in the console under API Access; sent on every request.
  ref: authentication/startree-cortexdata-authentication.yml
multi_tenancy:
  header: database
  value: Workspace ID (e.g. ws_2kc8e2dnzzb0)
  detail: Required on multi-tenant / Free Tier clusters; omit on dedicated clusters.
idempotency:
  supported: false
  detail: No idempotency-key header/param is documented; the Query API is read-only SQL execution.
pagination:
  style: sql
  detail: Result windowing is done in SQL (LIMIT/OFFSET) rather than a wire pagination convention.
response_envelope:
  shape: resultTable
  fields: [resultTable.dataSchema.columnNames, resultTable.dataSchema.columnDataTypes, resultTable.rows]
  detail: Query responses return a resultTable with a dataSchema (column names/types) and a rows array.
versioning:
  ref: lifecycle/startree-cortexdata-lifecycle.yml
notes: >-
  Cross-cutting semantics derived from the StarTree Cloud Query API spec and the
  API reference introduction. Broader Pinot Controller endpoints (tables,
  segments, tenants, ingestion) are documented at docs.startree.ai/api-reference.