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: Udesk
providerId: udesk
created: '2026-07-21'
modified: '2026-07-21'
generated: '2026-07-21'
method: searched
tags:
  - Rate Limiting
  - Customer Service
description: >-
  Published rate limits for Udesk's API families. The Open API (v2) and
  ServiceGo document a default of 60 calls per minute per tenant; the KCS
  knowledge SDK API documents a server-side Guava RateLimiter at 8 requests
  per second (100 ms token-acquisition timeout). No rate-limit response
  headers or 429 semantics are documented.
sources:
  - https://www.udesk.cn/doc/apiv2/intro/
  - https://www.udesk.cn/doc/sercive/general-instructions/
  - https://www.udesk.cn/doc/kcs/kcs-intro/
limits:
  - name: Open API v2 (default)
    scope: tenant
    metric: requests_per_minute
    limit: 60
    timeFrame: minute
  - name: ServiceGo API (default)
    scope: tenant
    metric: requests_per_minute
    limit: 60
    timeFrame: minute
  - name: KCS SDK API
    scope: service
    metric: requests_per_second
    limit: 8
    timeFrame: second
policies:
  - name: Server-side only
    description: APIs are designed for server-side calling; client-side calls risk
      token leakage and are disallowed by the docs.
  - name: Anti-replay
    description: v2 requests must carry a timestamp within 5 minutes of server time
      and a nonce that is unique within 15 minutes.