llm-intelligent-public-opinion-analytics

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow crawls social media feeds, hot lists, and video comments from public platforms like Weibo and Bilibili, feeding untrusted external posts directly into an automated LLM sentiment analysis and alert pipeline.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill clones a GitHub repository as part of its runtime setup instructions, which constitutes an unofficial external operational dependency with weak provenance.