gate-news-eventexplain

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md execution workflow explicitly calls news_events_get_latest_events, news_feed_search_news, and news_events_get_event_detail to ingest public news/event content (and then the LLM must read and use that content to attribute causes), so untrusted third‑party information from public news sources can materially influence the agent's reasoning and outputs.