A Spring self-check asks where Cloud routing and model output stop, and where application-owned policy begins.
Review before answering
- Spring Cloud release trains: pin the framework set before adding a starter
- Spring Cloud Config import: decide whether missing remote settings stop startup
- Spring Cloud client retries: distinguish a safe read from an uncertain write
- Spring Cloud Circuit Breaker: make fallback obey the original data contract
- Spring Cloud Gateway tenant routing: authenticate at the edge and enforce again at the service
- Spring AI ChatClient request budgets: bound latency, tokens and work per caller
- Spring AI retrieval: apply the tenant filter before prompt assembly
- Spring AI prompt injection: treat retrieved text as data, not authority
- Spring AI tools: authorize each requested action after model selection
- Spring AI conversation memory: partition by verified tenant and caller
- Spring AI structured output: validate the parsed object before any write
- Spring AI evaluations: separate model quality from enforced safety contracts
Check your reasoning
Answer every question and inspect each explanation. This browser self-check is not a credential, a live Cloud integration test or a model evaluation.
Common Mistakes
A route is not identity, parsed JSON is not authorization, and a model refusal is not a data-access control.