xianyu-auto-fulfillment

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill templates and scripts explicitly place secrets (SECRET_KEY, keys from key pools, API_KEY, bot tokens, webhook URLs) into commands, agent-browser messages, curl headers and logs, which requires the agent/LLM to output those secret values verbatim and thus creates exfiltration risk.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). These URLs include a personal cloud share link (https://pan.baidu.com/s/XXXXXXXXXXXX) used as a download delivery target, which is a high-risk vector for distributing arbitrary or executable files; the other URLs are placeholders or official APIs and are not direct download sources here.

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

• Third-party content exposure detected (medium risk: 0.30). 该技能在运行时对闲鱼聊天页面进行浏览器快照读取（agent-browser snapshot），并用其中的文本/按钮（如“我已付款，等待你发货”与“去发货”）触发发货流程，属于可被订单/聊天内容作者（第三方买家）投喂的间接提示注入面。