specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
generated: '2026-09-12'
method: searched
source: >-
  The provider's own OpenAPI documents (https://connect.greenscreens.ai/openapi.yaml and the 11
  per-service specs), the Triumph Intelligence help center
  (https://intelligencehelp.triumph.io/en/), and live probes of api.greenscreens.ai — all 2026-09-12.
provider: Green Screens AI
providerId: green-screens-ai
created: '2026-05-04'
modified: '2026-09-12'
description: >-
  Green Screens AI (Triumph Intelligence) enforces rate limiting but publishes no numbers and no
  runtime signal. A 429 with the error code too_many_requests is declared on 46 of the 47 operations in
  the aggregate spec, so the limit is real and documented as a failure mode — but no limit value, no
  window, no burst allowance, and no response header is published anywhere. A client cannot tell how
  much budget it has left, and cannot tell when it may retry.
limit_count: 0
limits: []
throttling:
  enforced: true
  documented_status: 429
  documented_code: too_many_requests
  documented_message: Request limit exceeded
  operations_declaring_429: 46
  operations_total: 47
  evidence: >-
    Every service spec declares a 429 response with schema <Service>_Error and an example
    {"code":"too_many_requests","message":"Request limit exceeded"}.
headers:
  limit: null
  remaining: null
  reset: null
  retryAfter: null
  policy: null
  note: >-
    No RateLimit-*, X-RateLimit-* or Retry-After header is documented on any operation, and none was
    present on the live responses observed 2026-09-12 (HTTP 401 from
    POST https://api.greenscreens.ai/v3/prediction/rates, HTTP 400 from
    POST https://api.greenscreens.ai/v1/auth/token). Recorded as absent after checking, not unchecked.
responseCodes:
  throttled: 429
  quotaExceeded: 429
  serviceUnavailable: null
published_quotas:
- scope: batch upload
  metric: file size
  limit: 100 MB
  source: https://intelligencehelp.triumph.io/en/articles/8520083-what-s-new
  note: >-
    The only numeric ceiling the provider publishes anywhere. February 2026 release notes: "Removed the
    lane caps for both short term and long term batch uploads. The only limit now is a file size of
    100 MB." This is a product upload limit, not an API rate limit, and it is not declared in any spec.
policies:
- name: Backoff
  description: >-
    Derived guidance, not provider policy: with no Retry-After and no reset header, exponential backoff
    with jitter is the only safe response to a 429 from this API.
  method: derived
note: >-
  This file previously carried invented per-tier limits (10/100/1000 requests per minute across Free,
  Professional and Enterprise tiers, with monthly quotas) written by the 2026-05-04 bulk sweep and
  marked method: generated. The provider publishes no tiers and no numbers; those values were scaffold
  defaults and have been removed. limit_count is an honest 0.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com