agency-trend-researcher

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow is explicitly designed to ingest “automated monitoring across 50+ sources” including “social media trends,” “news coverage,” “blog posts, forums, community discussions,” and “public web content” (via tools like Google Trends/SEMrush/Ahrefs/SimilarWeb), which are outsider-authored free text that can be fetched/scraped and then fed into the agent’s LLM context for analysis.