browser-use

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

• Third-party content exposure detected (high risk: 0.85). The skill is a browser automation framework that, at runtime, navigates to user-specified URLs and extracts page text/content (e.g., via page.goto(...) followed by page.textContent(...) / $$eval(...) / screenshot-to-vision), so outsider-authored free text from arbitrary web pages can be ingested into the agent’s LLM context.