landing-page-optimizer

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly asks for a page URL or screenshots and is used to "diagnose a page URL" (SKILL.md: "Ask for these inputs first
• URL or screenshots" and "Requests to diagnose a page URL"), which means the agent is expected to ingest and interpret content from public third‑party websites—an untrusted source that could carry indirect prompt-injection.