browser

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

• Third-party content exposure detected (high risk: 1.00). The skill's CLI explicitly navigates to arbitrary public URLs and extracts/interprets page content (see SKILL.md commands and EXAMPLES.md workflows using techcrunch.com, google.com, and other sites), so untrusted, user-generated third‑party webpages are fetched and their content can directly influence actions and subsequent tool use.