md-to-zhihu

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

• Third-party content exposure detected (high risk: 0.70). The skill's publish.py explicitly fetches and navigates public Zhihu pages and APIs (e.g., requests to https://www.zhihu.com/api_v4/... and Playwright navigation to https://www.zhihu.com/signin and https://zhuanlan.zhihu.com/write) and reads their responses/DOM (topic search results, login state, page content) to drive login, topic resolution, and publishing, exposing the agent to untrusted, user-generated third‑party content that can influence actions.