e.g. a screenshot, scrolling or not, that I've saved) and parse it (via OCR) for details in it.