Hacker News new | ask | show | jobs
by sheeshkebab 80 days ago
Obfuscated ts/js code is not machine code to begin with, so not sure what’s the big deal.

Also, not sure why anthropic doesn’t just make their cli open source - it’s not like it’s something special (Claude is, this cli thingy isn’t)

3 comments

> not sure why anthropic doesn’t just make their cli open source

They don't want everyone to see how poorly it's implemented and that the whole thing is a big fragile mess riddled with bugs. That's my experience anyway.

For instance, just recently their little CLI -> browser oauth login flow was generating malformed URLs and URLs pointing to a localhost port instead of their real website.

I don't think you really need to look at the source code to understand that it's probably been, let's say, written with a heavy help from Claude itself
Look at the gemini-cli.

Pretty sure it will look like that

I really don't think they care that much, but it's a tight race and gives them a slight edge over other labs building harnesses, since they are in the lead.
Browse through codex and think if anyone cares about the quality of the code before Open sourcing it.
caught that too a few weeks ago, couldn’t log in for a few hours either. I did a double take at the localhost when it loaded up in my browser haha
" - it’s not like it’s something special (Claude is, this cli thingy isn’t)"

How do you know? Have you checked the source?

Do you know how exactly context is created, memory files, skills? Subagents created with tasks?

I don't, but am checking right now. Then I will judge.

bc I build stuff like this myself - it doesn’t take anything to build a wrapper client around a good llm, including using another llm.
So .. the stuff you build yourself, you point it to claude then it runs just as productive as Claude CLI? Did you try?
Yes
Naming conventions can reveal a lot about how teams internally are thinking about roadmap and product decisions.

That cannot be reversed when obfuscated.