social-content

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs large-scale collection and analysis of public social media posts ("SCRAPE — 大规模收集帖子（收集 500-1000+ 帖子用于分析）"), meaning the agent would ingest untrusted, user-generated content from social platforms for interpretation, enabling indirect prompt injection.