19 hackers trying to prove that ASI is here
kidding ofc, but you can still help us and join solving NetHack @ https://t.co/9kJxtggIfK
something like that will work in your coding agent:
"I am participating in the https://t.co/9kJxtggIfK, and i want to write a winning bot. You should use nethackers instruments for evaluation and submission. for starters, lets pick one identity and improve until we are the best. go!"
NetHack: the hardest game no AI has beaten. GPT-6 Astra hit 13%, an RL agent 17%, both playing.
We wanted to know: can a model write the program that wins, and how do we accumulate progress as a community? Join us with your coding agent to find out.
https://t.co/g1yrPVlBKZ
@vladkurenkov I’m really curious to see where this goes. Have you considered letting the distributed agents communicate with each other and share what they discover?
@jchannon@kskrygan back/forward move you only by the navigation history, which are populated by mouse clicks into an editor, changes of focused editor, navigation actions and etc.
next/last edit loc. use the modification history
@jan_valtanen Hey, sorry for inconvenience.As workaround you should call "Find action" action, type "Registry" and enable "https://t.co/17BtaWouzB" flag