Summary
We are introducing organization-scoped HTTP rate limits (requests per minute, RPM) on selected Public API routes that are implemented in assignment-service and exposed through the Public APIsurface (as documented on the Developer Center / in the SDKs).
Behavior change :Under normal load, requests behave as today. When a caller exceeds the configured limit for an organization, the service may respond with HTTP 429 (Too Many Requests) with standard retry guidance (backoff, respect Retry-After when present). Integrations that assume unlimited throughput or that do not handle 429 may need to be updated to retry with backoff.
What is not changing: We are not removing or renaming resources, changing URL paths, or changing successful response schemas for these operations. Limits are enforced behind feature toggles and rolled out incrementally by environment(for example, Dev/Test first, then production regions in phases) so we can observe traffic and tune limits.
Affected areas (high level):Queue, agent, queue member, conversation, and related admin/operational endpoints that already appear in our public contract; a full list of paths, HTTP methods is listed below.
Effective Date
Monday, August 10, 2026
Details
Rate limits are being introduced to protect platform stability and ensure fair usage across customers.
Several Assignment Service endpoints support high-volume automation and operational workloads. Without throttling controls, excessive traffic from a single organization can impact latency and availability for other tenants sharing the same infrastructure.
Applying organization-level rate limits:
- Reduces noisy-neighbor effects
- Improves platform stability and predictability
- Supports capacity planning
- Aligns these endpoints with standard Genesys Cloud throttling behavior using HTTP 429 responses
The phased rollout approach allows traffic patterns to be evaluated and thresholds adjusted where necessary before full production deployment.
Customer Impact
Who may be affected:Customers and partners who call the affected Public API routes at high sustained volume or with tight retry loops may begin to see HTTP 429 (Too Many Requests) when their traffic for an organization exceeds the configured requests-per-minute limit.
What changes in practice:Successful responses (2xx) and error responses for invalid input (4xx other than 429) are unchanged when under the limit. When throttled, the client should back off and retry; if your integration does not handle 429 today, you may see intermittent failures until the client is updated.
Who is unlikely to be affected: Typical interactive or moderate-volume usage that stays below the published limits should behave as before.
Mitigation / rollout:Limits are applied per organization, behind feature toggles, and enabled incrementally so we can monitor traffic and adjust limits where needed. We will align Developer Center documentation with 429 behavior and recommended client patterns where applicable.
Impacted Resources
Queue
- PUT /v2/organizations/{orgId}/queues/bulk
- POST /v2/organizations/{orgId}/queues
- PUT /v2/organizations/{orgId}/queues/{queueId}
Agent
- PUT /v2/organizations/{orgId}/agents/{agentId}/queues
- PUT /v2/organizations/{orgId}/agents/{agentId}/queues/{queueId}
- PUT /v2/organizations/{orgId}/agents/{agentId}/status
Wrap-up Codes
- GET /organizations/{orgId}/wrapupCodes
Queue Members
- GET /v2/organizations/{orgId}/queues/{queueId}/users
- POST /v2/organizations/{orgId}/queues/{queueId}/users
- PUT /v2/organizations/{orgId}/queues/{queueId}/users
- PUT /v2/organizations/{orgId}/queues/{queueId}/users/{memberId}
- DELETE /v2/organizations/{orgId}/queues/{queueId}/users/{memberId}
Conversation
- PATCH /v2/organizations/{orgId}/conversations/{conversationId}
- POST /v2/organizations/{orgId}/conversations/{conversationId}/assign
Issue References
- AS-4497 — Queue Controller
- AS-4498 — Agent Endpoints
- AS-4394 — Wrap-up Code Query
- AS-4500 — Queue Member Operations
- AS-4502 — Conversation Controller
- AS-2886 — Queue Create / Update
Contacts
@Raghavendra Kopalle Please reply to this announcement with any questions. This helps the wider developer community benefit from the discussion. We encourage you to use this thread before contacting the designated person directly. Thank you for your understanding.