this would be really handy. i just tested, and after indexing, Cursor gave me the classic “I’m sorry, but as an AI, I currently don’t have the ability to access external URLs or documents outside of the provided context. Therefore, I can’t retrieve or analyze the content of the paper at the provided URL. I recommend manually reviewing the conclusion of the paper.”
re: how to do it:
you might have to do something on the scraper side to get the pdf file, and you might want to use something like unstructured.io to take care of the embedding.