Currently metadata parsing will scrape HTTP or HTTPS images, and https://aframe.io/aframe-registry/ throws Mixed Content warnings as a result.