ai-newsletters

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

• Third-party content exposure detected (high risk: 0.80). The skill's "Fetch feeds" workflow explicitly uses WebFetch to ingest public RSS feeds (https://bullrich.dev/tldr-rss/ai.rss and https://rss.beehiiv.com/feeds/2R3C6Bt5wj.xml), extracting titles/links/descriptions that the agent reads and interprets, exposing it to untrusted third-party content.