assistant-ui-streaming

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

• Third-party content exposure detected (low risk: 0.10). The skill’s described runtime workflow only decodes text that is streamed back from a response to the client/server (e.g., via AssistantStream.fromResponse(response, new DataStreamDecoder()) or useLocalRuntime yielding chunks from fetch("/api/chat")), not outsider-authored free text from an external user-controlled feed that the workflow automatically selects without prior targeting.