yuhua-perspective

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

• Third-party content exposure detected (high risk: 0.80). The skill's Agentic Protocol explicitly requires performing "余华式研究 (如需要)" in Step 2 and then states "基于 Step 2 获取的事实" (SKILL.md), and the included references/research/*.md list and link to many public URLs (e.g., ncpssd.cn, 163.com, baike.baidu.com), indicating the agent is expected to fetch and interpret open/public third‑party content that can influence its outputs and actions.