At the moment we only support a small number of files that contain simple magic byte strings, being able to parse the magic format used by services such as file would allow detection of many more files.