Simon Willison has a bunch of examples of scraping sites with GHA and storing the results in a repo. But you can use the same technique without the storing part if need be.