retellai-hello-world

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

• Third-party content exposure detected (high risk: 0.90). The skill creates a live voice agent tied to an LLM (Step 2's response_engine) and initiates phone/web calls (Step 3/4) whose incoming caller audio and resulting callDetail.transcript (Step 5) are untrusted, user-generated inputs that the agent reads and uses to drive its responses, enabling indirect prompt injection.