article-writer

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md workflow (Step 3: Material Collection) explicitly instructs using the content-planner's public wechat_search.js and a "web-search" skill to collect reference materials from the open web/WeChat articles, which the agent is expected to read and use to shape outlines and write content, so untrusted third‑party pages could indirectly inject instructions.