logoalt Hacker News

elxrtoday at 10:32 AM1 replyview on HN

A bit too vague to be useful advice don't you think?

Why not show some actual examples of these agents actually doing what you describe. How exactly would you set up an agent to simulate a user?


Replies

crabmuskettoday at 11:39 AM

To me it sounds like one way to do this would be to have LLMs write Cucumber test cases. Those are high level, natural language tests which could be run in a browser.