browser-rendering

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

• Third-party content exposure detected (high risk: 0.90). This skill fetches and renders arbitrary public web pages — e.g., REST API endpoints like /content, /scrape and the Workers examples that read a ?url parameter and call page.goto(url) and page.content()/page.$eval to extract text — so the agent will consume untrusted, user-provided third‑party content that could carry indirect prompt injections.