vector-composition

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

• Third-party content exposure detected (high risk: 0.90). The skill directly queries Scry's public corpus (e.g., scry.mv_posts, scry.mv_twitter_threads, scry.mv_hackernews_posts and other scry.mv_* views) to retrieve user-generated/open-web content which the agent is expected to read and use in downstream workflows (rerank, research-workflow), so untrusted third‑party content can materially influence tool use and decisions.