fetching-ai-news

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (SKILL.md and the MCP entry run_ai_news_digest in references/src/mcp_server.py) explicitly fetches and ingests content from public third‑party sources — Hacker News (references/src/sources/hackernews.py, user-generated), Google News RSS (references/src/sources/googlenews.py), and arXiv RSS (references/src/sources/arxiv.py) — and instructs the agent to read, classify, summarize, and optionally act (send) on those items, so untrusted external content can influence subsequent decisions or tool use.