generated: '2026-08-13'
method: searched
source: >-
  https://docs.appgain.io/integrations/n8n-integration/ +
  https://docs.appgain.io/Omni-channels/WhatsApp%20Lite/gettingStarted/ +
  https://docs.appgain.io/Marketing%20Automation/gettingStarted/ +
  https://headwayapp.co/appgain-changelog + live header probes 2026-08-13
limit_count: 0
api_rate_limits_published: false
note: >-
  Appgain publishes NO numeric API rate limit and returns NO rate-limit response
  headers. Its own integration guide instructs developers to "respect API rate limits
  and implement delays" and to "implement exponential backoff" on rate limiting —
  while never stating a limit, a window, or a status code. limit_count is 0 because
  there is no published API limit to record. The channel-level SEND caps and
  platform-side throttles below are real published numbers, but they are delivery
  policy on messages, not request limits on the API, so they are kept in a separate
  block rather than counted as rate limits.
response_headers:
  method: probed
  probed: '2026-08-13'
  rate_limit_headers: none
  observed_on_all_api_hosts: [x-api-version, x-kong-upstream-latency, x-kong-proxy-latency, via, access-control-expose-headers]
  retry_after: false
  status_on_exhaustion: undocumented
  note: >-
    The API sits behind Kong (via: kong/0.11.2), which supports rate-limiting plugins
    and standard headers, but none are emitted on the responses observed. An agent has
    no runtime signal to pace itself with — no remaining count, no reset time, no
    Retry-After.
channel_send_limits:
- channel: WhatsApp Lite
  scope: per connected number, per day
  limit: 300
  unit: messages
  guidance_range: 100-300 messages per day
  source: https://headwayapp.co/appgain-changelog
  note: >-
    Raised to 300/day with "adaptive pacing" in the WhatsApp Lite 1.0 release. The
    docs recommend staying at 100-300/day to avoid WhatsApp spam detection. Exceeding
    it risks the number being blocked by WhatsApp, for which Appgain states there is
    no refund.
- channel: WhatsApp Lite
  scope: platform-side throttle
  limit: adaptive
  unit: batching pause
  window: 30 minutes
  source: https://headwayapp.co/appgain-changelog
  note: '"Compliance Guard — adaptive 30-minute batching pauses automatically if Meta spam signals rise."'
- channel: WhatsApp Business API
  scope: per account
  limit: none
  source: https://appgain.io/whatsapp
  note: 'Appgain''s own comparison table states the Business API tier has "No Daily Limits" (Meta''s messaging tiers still apply upstream).'
platform_throttles:
- name: Smart Anti-Spam
  scope: per user, per automation event, per day
  limit: 1
  unit: message
  source: https://docs.appgain.io/Marketing%20Automation/gettingStarted/
  note: >-
    The automation engine will not send the same journey message to the same user more
    than once in a day even if the trigger fires repeatedly, and stops the sequence if
    the user clicks the CTA on an earlier message. This is a server-side delivery
    guard on journeys — it does not de-duplicate direct POST /{projectId}/send calls
    and is not an idempotency contract.
documented_guidance:
- {text: 'Rate Limiting: Respect API rate limits and implement delays', where: n8n integration guide, best practices}
- {text: 'Rate Limiting: Implement exponential backoff', where: n8n integration guide, common issues}
gaps:
- No published request-per-second/minute/hour limit for any endpoint.
- No 429 handling documented; no Retry-After.
- No per-key or per-project quota published.