generated: '2026-08-12'
method: probed
source: >-
  live anonymous requests to https://id.hadron.ad.gt/api/v1/pbhid ,
  https://seg.hadron.ad.gt/api/v1/rtd , https://analytics.hadron.ad.gt/api/v1/analytics ,
  https://uid2.hadron.ad.gt/v1/hadron.json (2026-08-12) ;
  https://docs.prebid.org/dev-docs/modules/userid-submodules/hadron.html ;
  https://docs.prebid.org/dev-docs/modules/hadronRtdProvider.html

limit_count: 0

summary: >-
  Audigent publishes no rate limits and its public endpoints emit no rate-limit
  headers. Full response headers were captured from all four live public hosts on
  2026-08-12: not one carries X-RateLimit-*, RateLimit-*, Retry-After or any
  quota signal, and the Prebid documentation pages that serve as Audigent's only
  developer reference say nothing about throttling, quotas or 429 handling. An
  agent calling this surface has no runtime signal to back off on, and no
  documented number to plan against. Recorded as an honest zero.

limits: []

response_headers_observed:
  rate_limit_headers_present: false
  checked_for:
  - X-RateLimit-Limit
  - X-RateLimit-Remaining
  - X-RateLimit-Reset
  - RateLimit-Limit
  - RateLimit-Remaining
  - RateLimit-Reset
  - RateLimit-Policy
  - Retry-After
  hosts:
  - host: id.hadron.ad.gt
    url: https://id.hadron.ad.gt/api/v1/pbhid?partner_id=0&_it=prebid&t=1&src=id&domain=example.com
    http_status: 200
    content_type: application/json
    headers_returned: [date, content-type, content-length, allow, access-control-allow-headers, access-control-allow-methods, access-control-allow-origin, server, cf-ray]
    rate_limit_headers: []
  - host: seg.hadron.ad.gt
    url: https://seg.hadron.ad.gt/api/v1/rtd?partner_id=0&_it=prebid&t=1
    http_status: 200
    content_type: application/json
    headers_returned: [date, content-type, content-length, server, access-control-allow-origin, allow, cf-cache-status, access-control-allow-headers, access-control-allow-methods, cf-ray]
    rate_limit_headers: []
  - host: analytics.hadron.ad.gt
    url: https://analytics.hadron.ad.gt/api/v1/analytics
    http_status: 200
    content_type: application/json
    headers_returned: [date, content-type, content-length, access-control-allow-origin, allow, access-control-allow-headers, server, cf-ray]
    rate_limit_headers: []
  - host: uid2.hadron.ad.gt
    url: https://uid2.hadron.ad.gt/v1/hadron.json
    http_status: 200
    content_type: application/json; charset=utf8
    headers_returned: [date, content-type, server, vary, access-control-allow-credentials, access-control-allow-headers, access-control-allow-methods, access-control-allow-origin, cf-cache-status, cf-ray]
    rate_limit_headers: []

exhaustion_behaviour:
  status_code: unknown
  note: >-
    Not established. No 429 was observed and no throttling policy is documented.
    This pipeline does not load-test a provider's production ad-serving
    infrastructure to find the ceiling, so the exhaustion behaviour is recorded as
    unknown rather than guessed.

edge:
  provider: Cloudflare
  note: >-
    All four hosts sit behind Cloudflare (server: cloudflare, cf-ray present), so
    Cloudflare-level bot management or DDoS protection may throttle abusive
    callers, but that is edge behaviour and not a published API rate limit. No
    Cloudflare challenge was returned to any anonymous probe.

documentation:
  published: false
  pages_checked:
  - {url: 'https://docs.prebid.org/dev-docs/modules/userid-submodules/hadron.html', http_status: 200, mentions_rate_limits: false}
  - {url: 'https://docs.prebid.org/dev-docs/modules/hadronRtdProvider.html', http_status: 200, mentions_rate_limits: false}
  - {url: 'https://docs.prebid.org/dev-docs/analytics/audigent.html', http_status: 200, mentions_rate_limits: false}
  note: >-
    Audigent hosts no developer documentation of its own; audigent.com soft-404s
    every unknown path (see well-known/audigent-well-known.yml), so no
    Audigent-hosted rate-limit page could exist and be discoverable anyway.

client_side_throttling:
  note: >-
    The only call-frequency control on this surface is client-side and lives in
    Audigent's own Prebid modules: hadronIdSystem caches the resolved ID in
    localStorage under `auHadronId` and hadronRtdProvider caches segment data
    under `auHadronRtd`, so a browser does not re-call on every auction. That is a
    caching convention, not a server-enforced limit.
  source: conventions/audigent-conventions.yml

remedy: >-
  Audigent could emit IETF draft RateLimit-Limit / RateLimit-Remaining /
  RateLimit-Reset headers on the Hadron endpoints and document a per-partner
  quota. Today an integrator has no way to know whether they are near a ceiling
  until traffic is silently dropped.