twilio-agent-augmentation-architect

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly ingests live call audio and transcripts via Conversation Intelligence (see SKILL.md Level 1/Level 2: "Live STT during calls → transcripts available" and "Conversation Intelligence attaches to live conversations → runs operators... → your backend pushes prompts to agent UI"), which are untrusted user-generated inputs that can drive Next-Best-Response suggestions and routing decisions and thus could enable indirect prompt injection.