generated: '2026-07-18'
method: derived
source: openapi/cuein-insights-openapi.json, openapi/cuein-answers-openapi.json
authentication:
  style: api-key
  header: x-api-key
  scheme: ApiKeyAuth
  ref: authentication/cuein-authentication.yml
versioning:
  scheme: uri-path
  current: v1
  evidence: servers use https://api.cuein.ai/insights/v1 and https://api.cuein.ai/answers/v1
pagination:
  style: cursor
  request_params:
    - nextToken   # token for next batch lookup (empty returns first page)
    - limit       # required; default system limits exist
    - timeframeFrom
    - timeframeTo
  response_fields:
    - meta        # ConversationBulkResponse.meta carries paging metadata
  applies_to:
    - openapi/cuein-insights-openapi.json#Retrieve insights for multiple conversations
idempotency:
  supported: false
  note: No idempotency-key header or parameter is documented in the OpenAPI.
request_tracing:
  supported: false
  note: No request-id / correlation header documented.
error_envelope:
  style: http-status
  problem_json: false
  ref: errors/cuein-problem-types.yml
rate_limit_signaling:
  documented: false
  ref: rate-limits/cuein-rate-limits.yml
  note: >-
    No 429 response, Retry-After, or X-RateLimit-*/RateLimit-* header appears in either
    published spec or anywhere in the developer hub. The only volume control expressed
    is a per-request `limit` query parameter whose description says "default system
    limits exist" without publishing them.
discovery:
  api_catalog: https://developer.cuein.ai/.well-known/api-catalog
  api_catalog_spec: RFC 9727
  llms_txt: https://cuein-api.readme.io/llms.txt
  ref: well-known/cuein-well-known.yml
metadata:
  additional_ids: ConversationResponse.additionalId (AdditionalIdKeyValue[] idKey/value pairs)
  source_metadata: ConversationResponse.sourceMetadata