generated: '2026-08-13'
method: searched
source: https://developers.scrunch.com/llms.txt
limit_count: 0
limits: []
response_headers: []
exhaustion_status: null
note: >
  Scrunch publishes no rate limits for the Data API. The full developer documentation
  index (developers.scrunch.com/llms.txt, 108 pages) contains zero occurrences of
  "rate limit", "throttle", or "429", and neither the live OpenAPI served at
  https://api.scrunchai.com/v1/openapi.json nor the docs-derived spec declares a 429
  response or any X-RateLimit-*/RateLimit-* response header on any operation. An honest
  zero: limits may be enforced server-side but nothing about them is documented, so an
  agent has no runtime signal to read.
pagination_caps_not_rate_limits: >
  The docs DO publish per-request result caps, which are pagination bounds rather than
  rate limits and are recorded in conventions/scrunch-ai-conventions.yml: query limit
  default 50000 / max 90000 rows; list endpoints default limit 100; listSignals default
  50 / max 200; MCP list_responses default 10 / max 100.
docs_checked:
- url: https://developers.scrunch.com/api-reference/introduction
  status: 200
- url: https://developers.scrunch.com/getting-started/authentication
  status: 200
- url: https://developers.scrunch.com/llms.txt
  status: 200
x-evidence:
- url: https://developers.scrunch.com/llms.txt
  status: 200
  fetched: '2026-08-13'
  finding: no rate-limit documentation in the complete docs index