wechat-title

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

• Third-party content exposure detected (low risk: 0.10). 工作流通过 fetch_title.py 向 doubaoya.com 的接口 POST https://doubaoya.com/api/apis/gongzhonghao/gongzhonghao-coze-cover/call 按用户给的 keyword 拉取并读取返回的标题数据（data.items[].item.title 等），其中自由文本内容由外部服务（doubaoya.com）生成/提供而非工作流主动选择的“外部用户提交文本”。

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill's runtime script calls the remote API https://doubaoya.com/api/apis/gongzhonghao/gongzhonghao-coze-cover/call to fetch JSON title data that is injected into the agent's context and directly controls the generated prompts/output, and this endpoint is required for the skill to operate.