page-audit

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

• Third-party content exposure detected (high risk: 1.00). This skill explicitly fetches the provided URL and performs Google searches to fetch and read top competitor pages as part of its required workflow ("Step 1: Fetch and Read the Page" and "Step 3: Research the SERP" in SKILL.md), so it ingests untrusted public third-party web content that directly influences its decisions and recommendations.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches and renders the user-provided audited URL (and the top-10 competitor URLs) at runtime and injects that remote page content into the agent's context to drive its instructions and outputs, so the user-supplied audited URL (the runtime input URL) is flagged.