wangjianshuo-perspective

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md Agentic Protocol (Step 2: "王建硕式研究") explicitly requires using WebSearch and similar tools to fetch and ingest web sources (see Step 2 and the referenced external sources in references/research/ such as NYT/TechNode/KQED and public blogs), and those fetched public/untrusted webpages are then read and used to drive the agent's research and subsequent actions, so the skill clearly consumes untrusted third‑party content that can influence behavior.