google-search-console

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

• Third-party content exposure detected (high risk: 0.80). The skill's documented actions (e.g., "Inspect URL" and "Run Mobile-Friendly Test" in the "Popular actions" section) invoke Google Search Console checks on public URLs via Membrane, meaning the agent will fetch/ingest arbitrary third‑party web content and use those results to drive decisions and follow-up actions.