cloudflare

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

• Third-party content exposure detected (high risk: 0.80). The skill's "Browser Rendering" section uses Puppeteer to navigate to an arbitrary URL taken from a request parameter (page.goto(url)) and extracts page content/text, which clearly fetches and ingests untrusted third‑party web pages that could contain instructions influencing the agent's behavior.