building-ai-agent-on-cloudflare

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

• Third-party content exposure detected (high risk: 0.95). The required runtime path builds the LLM prompt messages using data.content / userMessage from the connected client over WebSocket (outsider-authored chat content, e.g., anyone other than the operating user), which is then passed into this.env.AI.run(...) as messages.