summarize

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md "URL / Web Content" workflow explicitly directs the agent to "Fetch the content using available tools (web fetch, curl)" and extract main text from URLs/articles, which means it will ingest and act on arbitrary public third‑party web content that could contain untrusted/user-generated instructions.