weread-skills

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

• Third-party content exposure detected (medium risk: 0.30). 该技能需要在用户选择后调用微信读书接口读取/展示可包含自由文本的内容字段（如从 /user/notebooks、/book/bookmarklist 的 markText、/review/list/mine 的 review.content 等返回并直接用于回答/导出），因此在运行时会摄入由平台用户/他人作者撰写的文本内容。

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

• Potentially malicious external URL detected (high risk: 0.90). The skill makes runtime calls to the gateway API at https://i.weread.qq.com/api/agent/gateway and explicitly requires obeying server-returned "upgrade_info.message" (remote content) which can directly control the agent's instructions, so this external URL is a runtime dependency that can change agent behavior.