agentation-self-driving

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly opens arbitrary web pages with agent-browser --headed open <url> and then snapshots and evals page DOM (e.g. snapshot -i, document.querySelector(...).textContent) to read and annotate site content, which exposes the agent to untrusted public third‑party webpages and their user-generated content.