library-architecture

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

• Third-party content exposure detected (high risk: 0.70). The skill's Content Rules explicitly require collecting material "from outside your domain" and the Constraints instruct "MUST save screenshots, not bookmarks," which implies the agent will fetch and ingest third-party/public web content as part of its workflow where that content can influence curation and tooling decisions.