generated: '2026-08-05'
method: searched
source: https://public-api.tryriot.com/openapi
docs: https://docs.tryriot.com/
description: >-
  Rate-limit signalling for the Riot Public API. Riot documents that rate limiting exists, how it is scoped
  and what happens when it is exceeded, but does NOT publish the numeric limits.
rate_limits:
- api: Riot Public API
  scope: per authentication key
  scope_note: All requests made with the same x-api-key share the same limit.
  algorithm: fixed time interval window
  limit_count: null
  window: null
  published: false
  published_note: >-
    "Specific rate limits are defined and managed by the technical team." No numeric quota, window length or
    per-endpoint limit is published anywhere on the public surface.
  exceeded:
    status: 429
    error_code: too_many_requests
    error_title: Too Many Requests
    schema: RateLimitExceededErrorResponse
    behavior: >-
      Further requests return 429 until the next interval begins.
  headers:
    retry_after: not documented
    ratelimit_limit: not documented
    ratelimit_remaining: not documented
    ratelimit_reset: not documented
    note: >-
      No RateLimit-* / X-RateLimit-* / Retry-After response headers are declared in the OpenAPI or described
      in the docs, so a client cannot read its remaining budget or the reset time — it can only observe 429.
  applies_to: all API endpoints
  declared_on_operations: 24
  declared_on_operations_note: >-
    Every /v1/* operation declares a 429 response referencing the shared RateLimitExceededErrorResponse.
    The /scim/* operations do not declare 429.
gaps:
- No numeric limit is published — an integrator cannot size a backoff or a batch job in advance.
- No rate-limit response headers are documented, so budget must be discovered by hitting 429.
- No Retry-After is declared on the 429 response, leaving retry timing to the client's guess.
cross_links:
  conventions: conventions/riot-conventions.yml
  errors: errors/riot-error-codes.yml