vercel-sandbox

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

• Third-party content exposure detected (high risk: 0.90). The skill's Flow explicitly instructs the agent to "Navigate to target URL" (and lists "Target URLs" in Environment Variables) and then to "extract" and perform actions (click, type) on page content, so arbitrary/untrusted web pages can be ingested and materially influence the agent's actions.