specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
provider: Knock
providerId: knock-app
created: '2026-05-25'
modified: '2026-05-25'
reconciled: true
tags:
  - Notifications
  - Rate Limiting
  - Quotas
description: |
  Reconciled rate limits for the Knock API. Each endpoint is assigned a tier (T1-T5).
  Limits are scoped by environment when using public or private API keys; when using
  signed user tokens (enhanced security mode), limits are additionally scoped per user.
  Knock returns HTTP 429 when a limit is exceeded.
sources:
  - https://docs.knock.app/api-reference/overview/rate-limits
  - https://docs.knock.app/api-reference/overview/batch-rate-limits
responseCodes:
  throttled: 429
  quotaExceeded: 429
algorithm: tiered-token-bucket
scope: |
  Per environment for API-key requests; per environment + signing user for signed-user-token
  (enhanced security mode) requests.
authoritativeCredentials:
  - publicApiKey
  - secretApiKey
  - signedUserToken
tiers:
  - tier: T1
    rate: 1
    unit: request
    timeWindow: second
    description: One request per second. Reserved for the most write-heavy or expensive
      endpoints.
  - tier: T2
    rate: 5
    unit: request
    timeWindow: second
  - tier: T3
    rate: 60
    unit: request
    timeWindow: second
  - tier: T4
    rate: 200
    unit: request
    timeWindow: second
  - tier: T5
    rate: 1000
    unit: request
    timeWindow: second
    description: Highest-throughput tier — used for read-heavy and bulk endpoints.
notes:
  - Rate-limit tiers are assigned per endpoint, not per plan.
  - Batch endpoints have additional per-item caps documented under "Batch rate limits".
  - For high-volume workloads, Knock will provision a custom tier on request.