It would be nice to identify the specific file that is too large to synchronize with the API. It would be a copy of the maximum file sizes in the CLI to do the character length check locally first.
Or update the API to show which files are too large (not just the hash check).