agent-native-architecture

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly recommends adding an HTTP request primitive (fetch_url) as a core primitive in references/from-primitives-to-domain-tools.md ("Start with Pure Primitives" lists HTTP requests (fetch_url)), which allows the agent to fetch and read arbitrary third‑party webpages that could influence its decisions and tool use.