openwa

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

• Third-party content exposure detected (high risk: 0.75). Outsider-authored free text can enter the LLM context via webhook delivery: the runtime ingests the HTTP request body (including fields like data.body from message.received) into the agent’s context when the agent processes incoming webhook/event payloads; this body is authored by external WhatsApp users (outsider) and is forwarded as JSON prose to the agent.