generated: '2026-08-13'
method: searched
source: >-
  https://www.hull.io/docs/connectors/outgoing-user-webhooks/,
  https://www.hull.io/docs/connectors/outgoing-account-webhooks/,
  https://www.hull.io/docs/reference/hull_node/api/,
  https://www.hull.io/pricing/
limit_count: 2
api: Hull HTTP API
summary: >-
  Hull publishes NO numeric rate limit for its inbound HTTP API and documents no
  rate-limit response headers. The only published, numeric limits are on the
  OUTBOUND side — the Outgoing Webhooks connectors, where the operator sets the
  request rate and concurrency and Hull enforces a hard batch-timeout budget.
  Hull markets "Intelligent Queuing, Adaptive Rate Limits" as a platform feature
  on every pricing tier, but the adaptive limits are never quantified in public
  docs. Inbound throughput is governed commercially instead, by the monthly
  incoming-request quota attached to each plan (see plans/hull-plans-pricing.yml).
response_headers:
  documented: false
  headers: []
  note: >-
    No X-RateLimit-*, RateLimit-* or Retry-After header is documented anywhere in
    the HTTP API reference, the status-codes page, or the Node client reference.
    An agent has no runtime signal to read.
exhaustion_status:
  documented: false
  note: >-
    429 is not listed among the status codes Hull documents for its own API
    (200/201/400/401/404/405/500/503). The Node Connector SDK does define a
    RateLimitError raised on `res.status === 429`, but that handler is for the
    THIRD-PARTY service APIs a connector calls, not for Hull's own API.
limits:
  - scope: per-webhook-endpoint (Outgoing User Webhooks / Outgoing Account Webhooks connector)
    setting: Requests rate limit
    window: 1 second
    limit: operator-configured
    description: The maximum number of requests the connector performs every 1 second.
    docs: https://www.hull.io/docs/connectors/outgoing-user-webhooks/
  - scope: per-webhook-endpoint (Outgoing User Webhooks / Outgoing Account Webhooks connector)
    setting: Requests concurrency
    window: n/a
    limit: operator-configured
    burst: >-
      Caps simultaneous open requests; with concurrency 2 the first two fire
      immediately and the third waits for either to complete.
    docs: https://www.hull.io/docs/connectors/outgoing-user-webhooks/
platform_timeout_budget:
  rule: "10 requests needs to fit in 25 seconds time window"
  detail: >-
    Hull's internal batching and timeout levels require the connector to process
    10 users in under ~20 seconds. With concurrency 1 the slowest webhook
    endpoint must respond in under ~2 seconds or Hull's timeout interrupts the
    data flow.
  receiver_requirement: >-
    The receiver must answer 200-204; anything else is recorded as
    outgoing.account.error. Redirects (301/302) are treated as errors, not
    followed.
  docs: https://www.hull.io/docs/connectors/outgoing-account-webhooks/
commercial_quota:
  metric: incoming requests per month
  ref: plans/hull-plans-pricing.yml
  values: {Team: 10000000, Business: 20000000, Business +: custom}
  overage: {Team: $190 per 1M, Business: $180 per 1M}