Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

As a long-time Cursor user, I just tried Claude Code for the first time two days ago and I found it:

- Burning tokens with constant incorrect command-line calls to read lines (which it eventually gets right but seemingly needs to self-correct 3+ times for most read calls)

- Writing the string "EOF" to the end of the file it's appending to with cat

- Writing "\!=" instead of "!="

- Charged me $7 to write like 23 lines (admittedly my fault since I forgot I kept "/model opus" on)

Minus the bizarre invalid characters I have to erase, the code in the final output was always correct, but definitely not impressive since I've never seen Cursor do things like that.

Otherwise, the agent behavior basically seems the same as Cursor's agent mode, to me.

I know the $7 for a single function thing would be resolved if I buy the $100/month flat fee plan, but I'm really not sure if I want to.



The Pro account is only $20/month and works with Claude Code.


I learned about it from this thread and will buy a month's worth to keep playing with it. (48 hours ago the documentation said it was only supported for Max.)


I received the e-mail exactly 15 hours ago:

   Hello,
   
   Your Pro plan just got way more powerful with three major upgrades previously available only to Max, Team, and Enterprise users. 
   
   Claude Code is now included
   
   Claude Code is a command line tool that gives you direct access to Claude in your >terminal, letting you delegate complex coding tasks while maintaining full control. You can now use Claude Code at no extra cost with your Pro subscription.



Not quite. You pay ala carte for it with API pricing.

The $100/mo max plan lets you use a Claude Code with a fixed bill. There’s some usage limits though.


They add Claude Code to the pro plan yesterday: https://x.com/_catwu/status/1930307574387363948


Ha! That’s what I get for hallucinating using two day old data.


I'm not sure an x.com link to a GIF really helps clarify the status of Claude Code on Pro plans. Here's the actual anthropic docs on it: https://support.anthropic.com/en/articles/11145838-using-cla...


> Burning tokens with constant incorrect command-line calls to read lines

See this sort of asinine behavior with cursor too sometimes although it's less grating when you're not being directly billed by the failed command line attempt. Also it's in a text editor it fully controls why is it messing around in the command line to read parts of files or put edits in, seems to be a weird failure state it gets into a few times a project for me.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: