why

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

• Third-party content exposure detected (medium risk: 0.30). This skill at runtime explicitly spawns investigators that query external MCP evidence sources (e.g., issue/ticket trackers, long-form docs, and real-time chat) and instructs investigators to read full tickets/docs/threads, which could include outsider-authored free text delivered through those systems.