whatsapp-cloud-api

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

• Third-party content exposure detected (high risk: 0.90). High: the webhook POST handler processes incoming WhatsApp customer text/media into message.text.body (outsider-authored user messages) and that userText is then fed into the Claude integration via getAIResponse(session, userText) / anthropic.messages.create(..., messages), so arbitrary customer free text reaches the LLM context at runtime.