Is there anyway some of these can be put in another ‘Legacy’ section to hide them away unless needed by the user?

----------回复的邮件信息----------
Dean Rie via Cursor - Community Forum notifications@cursor1.discoursemail.com 在 Fri,Mar 28,2025 4:06 PM写道：

| deanrie Community Helper
March 28 |

• | - |

Hey, open Cursor settings with Cmd/Ctrl + Shift + J, and on the General tab, you’ll see the logout button.

(Request ID: 2724fe3e-b638-4a62-af82-5a84365154cb)

I now have a message telling me that an update is available though (although im using the latest version from the website)

I’m not clicking that for now…

(Request ID: 3004f856-6920-443f-

and I still have some credit:
You’ve used 186 requests out of your 500 monthly fast requests quota.

Is there a new limit on conversation length? This is quite inconvenient, as the cursor should automatically manage conversation history or use specific code within the conversation. Instead, we are forced to manually create a new conversation.
Note: my last message is very short, this is due to long conversation history.

Thank you!

My chats are stuck on “generating” so it looks like whatever it is, is not resolved.

I’ve tried to start new chat, change models, restart cursor, update cursor and the usual debugging steps (remove cursor rules, limit context etc).

I occasionally get a reply but then the conversation hangs on “generating…”

I have credits left.

This bug is not resolved and is still happening (March 2025).

• I’m using the latest version of Cursor.
• Switching model does not help.
• If i start a new chat, 9/10 times i have the same problem, and the 1/10 times i get a response it goes back into an endless “generating…” loop after about 10 seconds.
• I have plenty of credits remaining.
• I’ve tried both my own API key and also using the standard Cursor setup (no custom api keys) - i have the same behaviour in both cases.
• I have a small project (< 2000 lines of code)
• i do have several cursor rules files, but if i remove those, i get the same issue,

I have tried switching restarting everything just in case that helps (it doesn’t)

If you want to try it anyway you can select it from Agent mode now - You will be charged at the premium rate though - To get around this you can use your own API key in settings, ignore the verify error - This is available in Ai Studio and is free - I think you have a 50 call per day limit though…

Basically, I tried a quick and easy ToDo app and it had errors - Not a good look!..

• Switching model does not help.

• Starting a new chat helps temporarily, as in - Chat starts to answer my question then after 5 seconds or so i get the endless “generating…” loop.

My questions

1. Does it count as a request when it updates and applies changes to a file?
2. Does using Claude 3.7 Sonnet over 3.5 use more requests?
3. Does using thinking mode, particularly for Claude 3.7, use more requests?

I think 3.5 is more deterministic and more reliable, but sometimes it can lack the creativity of 3.7.

@https://forum.cursor.com/t/agent-not-detecting-that-a-command-has-completed/65052/4 
solution? you have this bug. how to solve this?
run echo first

Solution:

Now let's try running it with a simpler command and redirecting both stdout and stderr:

python test.py 2>&1

cursor has been on a bender for two weeks, drinking too much of the good stuff. I use two paid accounts, it pulls off quick replies faster than a German yells “another beer!”.

The rules files are constantly ignored, except for the one I set in the “User Rules”. I got a little annoyed that the cursor was apologizing like a wuss while quickly burning my tickets. So I made the conversation a bit more fun, at least for me. If I’m going to lose time and money, I want to have a little something to say about how I do it

On a side note: In one rule, I requested that whenever he learned something important or I took the time to explain why certain things don’t work and show how to do it correctly, he should save that information in the .cursorrules file. It has done that - but as I said, it keeps ignoring that information. Proponents of the end-time belief could possibly interpret the write access of an AI on their own rulebook as one of the 7 signs

I actually tried some simple tasks where Cursor failed with Claude 3.7 sonnet (thinking) on Anthropic. Given no information, the LLMs made exactly the same mistakes. I wouldn’t expect it to be any different - but for a programming companion - context control is crucial. When Cursor started to fail to alter files on my PC, I began using Anthropic chat - which seems to have at least 4 times the contextual knowledge window - until the FIFO kicked out the important information as well.

My friend develops web apps with the cursor - and it’s still great what they produce. But whatever you put into automatic/intelligent context handling, but lately it
isn’t working for my projects anymore, involving multiple frameworks, programming languages, interpreters and shells/operating systems.

What I’d genuinely like to know and understand, just for the sake of curiosity, is what can possibly be the reason that a well-performing service has been dropping out like a cursor lately? I mean, the Anthropic APIs haven’t changed, the model is evidently static.
About the only conjecture I can imagine: a flawed optimization reducing token usage.