generated: '2026-08-14'
method: probed
source: live responses from https://api.avarra.ai on 2026-08-14
limit_count: 0
limits: []
headers: []
status_on_exhaustion: unknown
retry_after: false
note: >-
  No rate limits are documented (Avarra publishes no developer documentation at all)
  and none are signalled on the wire. The observed responses from api.avarra.ai carry
  no X-RateLimit-*, RateLimit- (RFC 9331 draft) or Retry-After header. An agent
  calling this API has no runtime budget signal and would have to discover throttling
  by being throttled. This is an honest zero, measured - not an unchecked field.
observed_response_headers:
- request: GET https://api.avarra.ai/health
  status: 200
  headers:
  - date
  - content-type
  - content-length
  - vary
- request: GET https://api.avarra.ai/v1/me
  status: 401
  headers:
  - date
  - content-type
  - content-length
  - vary
unrelated_signal:
  note: >-
    https://status.avarra.ai returns "ratelimit-policy: 1000; w=60", but that host is
    a Better Stack hosted status page and the header is Better Stack's, not Avarra's
    API. It is recorded here only so a later pass does not mistake it for an Avarra
    API limit.