ai-news-digest

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

• Third-party content exposure detected (high risk: 0.85). Required workflow runs scripts/fetch_ai_news.py which fetches RSS/Atom XML from enabled third-party URLs in references/sources.json (outsider-authored web content) and ingests their readable title/description/summary text into the generated digest/JSON that the agent then uses as LLM context.