agent-native-architecture

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

• Third-party content exposure detected (high risk: 0.90). This skill's runtime workflow and tests explicitly instruct agents to perform web searches and HTTP fetches (e.g., references/architecture-patterns.md "WebTools.webSearch(), WebTools.webFetch()", references/from-primitives-to-domain-tools.md listing "HTTP requests (fetch_url)", and references/agent-native-testing.md integration test "Research flow: request → web search → file creation"), meaning the agent will ingest open web / third‑party content (untrusted user-generated pages) and act on it as part of its tool-driven workflows.