analyze-stock

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs multiple subagents to perform WebSearch and WebFetch on public sites (e.g., "搜索…最新新闻", "使用 WebFetch 访问公司官网", and to visit news original URLs in Agent 2 and other agents), so the agent will fetch and interpret open/public third‑party content (news, websites, search results) that can materially influence analysis and subsequent actions.