visual-qa

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

• Third-party content exposure detected (medium risk: 0.65). SKILL.md Step 1/3 uses runtime browser tooling (browser_links discovery and browser_screenshot(url, ...)) to fetch arbitrary user-supplied or discovered web pages, whose page content can be ingested as readable text by the LLM during the screenshot/processing and thus constitutes outsider-authored free text from public web content.