llm-public-opinion-analytics

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

• Third-party content exposure detected (medium risk: 0.30). The public opinion analysis assistant scrapes social media and news feeds from external platforms (Weibo, Bilibili, Zhihu, etc.) and processes outsider-authored titles and content through an LLM.

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

• Potentially malicious external URL detected (high risk: 0.70). The repository clone URL https://github.com/hmmnxkl/LLM-Based-Intelligent-Public-Opinion-Analytics-Assistant.git is an arbitrary third-party personal GitHub repository providing the operational code and skill material, representing an unverifiable weak/unknown dependency.