generated: '2026-07-19'
method: searched
source: https://docs.endgame.io/api-reference/endpoints
docs:
- https://docs.endgame.io/api-reference/authentication
- https://docs.endgame.io/api-reference/endpoints
authentication:
  style: bearer
  header: Authorization
  formats:
  - api-key (eak_ prefix, Personal or Org-wide scope)
  - WorkOS M2M access token (client_credentials grant)
  ref: authentication/end-game-authentication.yml
idempotency:
  supported: false
  note: >-
    No idempotency-key header is documented for the public API. Thread creation is
    a POST that triggers asynchronous assistant generation; callers poll the thread
    status rather than retrying idempotently.
pagination:
  style: cursor
  request_params:
  - limit (integer 1-100, default 25)
  - cursor (opaque, from previous response nextCursor)
  response_fields:
  - nextCursor (opaque; null when no more pages)
  - truncated (true when org has more threads than the pagination window can walk)
  filtering:
  - accountId (optional filter on listThreads)
versioning:
  scheme: uri-path
  current: v1
  base_path: /api/v1
async_model:
  note: >-
    Assistant responses generate asynchronously. After createThread, poll
    getThread until status.state is `idle` (success) or `error`.
request_tracing:
  field: error.trace_id
  note: Every error envelope carries a trace_id to include in support requests.
error_envelope:
  ref: errors/end-game-error-codes.yml
  shape: '{ "error": { "code", "message", "trace_id" } }'
rate_limiting:
  ref: rate-limits/end-game-rate-limits.yml
  signal: HTTP 429 with error.code RATE_LIMITED
  default_quota: 10000 requests per organization per day (beta)