ce-agent-native-architecture

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly defines and uses tools that fetch and ingest open web and third-party content—e.g., web_search/web_fetch in references/architecture-patterns.md and references/agent-execution-patterns.md, a generic call_api/fetch_url tool in references/mcp-tool-design.md, and integration tests that expect web searches and Discord messages to be read and acted on—so the agent will read untrusted external content and use it to drive actions.