optimize-seo

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

• Third-party content exposure detected (high risk: 1.00). The skill's Step 1 and Example 3 explicitly require using WebFetch to retrieve and audit arbitrary URLs (e.g., "If the user provides a URL, read it" and "WebFetch the URL"), so the agent will fetch and interpret open/public web pages that may contain untrusted user-generated content which could influence its decisions and actions.