I image it depends on the book/article sources. Some books/articles are behind paywalls. Either way, automating this into markdown may need to be a custom webscraper. Sounds like a fun project!