If you're using Org-ref, and have used that to drill down into your pdfs then you always have the document context (and bibtex etc context) from within emacs. But I don't think that is the OPs use case, so apologies for the noise. Also careful use of unix find with pdfgrep can be handy.