seo-sxo

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches arbitrary public pages ("Step 1: Fetch the target URL via scripts/fetch_page.py") and performs live SERP/backwards analysis ("Step 2: Search Google for the target keyword (WebSearch)" extracting top-10 results, PAA, related searches, AI Overview, etc.), ingesting untrusted third‑party content which is then interpreted to drive classifications, recommendations and tool-driven actions—thus exposing the agent to potential indirect prompt injection.