generated: '2026-08-12'
method: probed
source: 'live anonymous requests against https://api.pebblepost.com and
  https://api.pbbl.co, plus a search of https://docs.pebblepost.com/ for any published
  limit'
# NO PUBLISHED LIMITS AND NO RUNTIME SIGNAL. Both facts were checked separately and
# both are negative: PebblePost documents no rate limit anywhere, AND neither live host
# returns a rate-limit header on an anonymous request. An integrator has no number to
# design against and no header to back off on.
limit_count: 0
documented: false
headers_observed: false
retry_after: false
status_on_exhaustion: null
docs: null
probes:
- host: api.pebblepost.com
  url: https://api.pebblepost.com/health
  status: 200
  runtime: Express
  response_headers:
  - date
  - content-type
  - content-length
  - x-powered-by
  - etag
  rate_limit_headers: []
  note: 'No X-RateLimit-*, no RateLimit-* (RFC 9331 style), no Retry-After. 12
    consecutive requests issued back to back all returned 200 with no throttling
    signal, so no limit could be observed either.'
  burst_test:
    requests: 12
    interval: back-to-back
    statuses_returned: [200]
    throttled: false
- host: api.pbbl.co
  url: https://api.pbbl.co/
  status: 403
  runtime: AWS API Gateway
  response_headers:
  - date
  - content-type
  - content-length
  - x-amzn-requestid
  - x-amzn-errortype
  - x-amz-apigw-id
  rate_limit_headers: []
  note: 'AWS API Gateway supports usage plans with throttle and quota settings, and
    returns 429 with {"message":"Too Many Requests"} when they are exceeded — but no
    such plan is observable anonymously, because every path is rejected at the
    authentication stage first with 403 MissingAuthenticationToken. Whether a quota
    exists behind credentials is unknown and is NOT asserted here.'
limits: []
x-gap:
- 'No rate limit is published and none is signalled at runtime. An agent or integrator
  cannot implement correct backoff: there is no limit to respect, no counter to read,
  and no Retry-After to obey. Emitting standard RateLimit-Limit / RateLimit-Remaining /
  RateLimit-Reset headers, or simply documenting the API Gateway usage-plan throttle
  that likely already exists, would close this with no new engineering.'