topview-skill

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and displays content from the public Topview service (e.g., board/task results and public avatar templates via references/board.md and references/product_avatar.md's list-avatars/list-categories, and even supports an internet search flag in references/video_gen.md --internet-search), which are user-generated/public resources the agent is expected to read and use to choose avatars/boards/models—so untrusted third‑party content can materially influence tool selection and next actions.