azure-ai-agents-persistent-java

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The runtime path only reads provider-authored SDK usage data and ingests user-supplied message text via client.createMessage(thread.getId(), MessageRole.USER, ...) and then processes it through client.createRun(thread.getId(), agent.getId()), which is an outsider-controlled free-text entry into the workflow.