finding-changelogs

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to fetch and evaluate public web pages, feeds, and raw markdown (e.g., "Fetch the page with WebFetch", feed discovery probes, well-known files, and scraping/crawl fallbacks) from arbitrary third‑party sites and GitHub, and to use that content to decide recommended ingestion methods—meaning untrusted external content is read and can change agent actions.