worker-classification

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow (步骤2 and the "研究连接器预检" / step 5) explicitly requires running legal research via configured connectors (e.g., yuandian MCP) and offers an option to "搜索网络" with results labeled "[联网检索——需复核]", which means it will fetch and interpret open/public web content as part of its required analysis.