Show HN: Ego lite – why our browser agent writes JavaScript not CLI commands
1–10 of 13 posts
Re: Show HN: Ego lite – why our browser agent writes JavaScript not CLI commands
#2Re: Show HN: Ego lite – why our browser agent writes JavaScript not CLI commands
#3Re: Show HN: Ego lite – why our browser agent writes JavaScript not CLI commands
#4smart move, when I was using agent browser,I have to watch it execute commands one by one, it’s really inefficient. using JS code to execute all commands at once is smart.
Re: Show HN: Ego lite – why our browser agent writes JavaScript not CLI commands
#5Re: Show HN: Ego lite – why our browser agent writes JavaScript not CLI commands
#6A browser that can run multiple agent-controlled sessions at the same time basically turns multiboxing from a chore into a one-click experience. My productivity may not improve, but my game progression definitely will.
Re: Show HN: Ego lite – why our browser agent writes JavaScript not CLI commands
#7Re: Show HN: Ego lite – why our browser agent writes JavaScript not CLI commands
#8smart move, when I was using agent browser,I have to watch it execute commands one by one, it’s really inefficient. using JS code to execute all commands at once is smart.
Why javascript ?
Re: Show HN: Ego lite – why our browser agent writes JavaScript not CLI commands
#9Earlier quoted context omitted.
Why javascript ?
I have read the repo, from what I can see is the thing used to connect the browser is writen in js, command and tool all stay in js makes the execution faster, make sense, but I m not sure, coz I haven’t read about their benchmark, could just a story lol
My point is here: There are many popular AI-Friendly PL, eg. Python, Rust ...
I also see many CDP-based framework which implemented by Python-Binding libs, They works very well.
I think javascript is not the only option.