ios-networking

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

• Third-party content exposure detected (high risk: 0.90). This skill's workflows explicitly perform HTTP requests and streaming connections to arbitrary external URLs and endpoints (see "Core URLSession async/await", Endpoint/urlRequest construction and the full client in references/urlsession-patterns.md, plus SSE and URLSessionWebSocketTask/Network.framework sections) and then decode and act on those responses (JSON, SSE, WebSocket messages), meaning it ingests untrusted third‑party content that could influence agent behavior.