telnyx-webrtc-client-android

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

• Third-party content exposure detected (high risk: 0.90). The skill consumes runtime messages from external parties (e.g., Telnyx socket messages via socketResponseFlow/ReceivedMessageBody and transcriptUpdateFlow, plus FCM push payloads parsed in FirebaseMessagingService) and explicitly uses that untrusted, user-generated content to drive actions like acceptCall/endCall/sendAIAssistantMessage, so third‑party content can materially influence the agent's behavior.