Hi! Would improve usability of the task greatly to be able to take base64 strings as that is what is usually provided from APIs when fetching pdfs.
