wechat-hot-write

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

• Third-party content exposure detected (medium risk: 0.30). 在必需工作流第2步中，脚本 scripts/hot_write.py 会根据用户给定的 keyword/start/end 向 doubaoya.com 的 POST https://doubaoya.com/api/apis/gongzhonghao/hot-article/call 拉取“同主题爆文”样本 JSON，并将 data（其中来自外部内容）直接输出给主 Agent 进行提炼。

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

• Potentially malicious external URL detected (high risk: 0.70). The runtime script posts to https://doubaoya.com/api/apis/gongzhonghao/hot-article/call to fetch JSON "爆文" samples which the main agent uses as the basis for prompts/instructions, so remote content can directly control the agent's outputs.