You can now plan with one model and implement with another. Build the plan locally or in the cloud.

We've improved the agent harness for all models. https://x.com/jediahkatz/statu...

Review the agent's changes across your codebase without needing to jump between individual files.

We've simplified the prompt input.

Cloud agents are now available in-editor. Switch from local → cloud to run agents while you're away from your computer.

A new confirmation modal to prevent accidentally closing Cursor. https://x.com/kr0der/status/19...

We've updated our web interface. https://x.com/WilliamHolmbe19/...

Finally, we've drastically improved the performance of using LSPs. Python and TypeScript LSPs are now faster by default. Memory is dynamically configured based on available RAM. We've also fixed a number of memory leaks and improved memory usage.

@cursor_ai But did you use Cursor for that?

@cursor_ai Did you use cursor for that?

@cursor_ai Give me the option to let it execute its own plan autonomously without having to reprompt it just to continue what it already knows it needs to do every 60 seconds—it's really annoying.

@cursor_ai When I click the "Add New MCP Service" button, it opens the previously installed MCP, and I cannot add a new MCP. This issue still hasn't been improved.

@cursor_ai finally feels less clunky for jumping between models without losing the flow.

@cursor_ai Dang. @cursor_ai just shipped 2.0 and only a week later, they ship a massive QoL update like it’s nothing.

@cursor_ai these are huuuuuge! BTW, there are some bugs in the plan preview & markdown view. I couldnt search using CMD + F, which is kinda annoying + it auto scrolls up/down when i switch files & come back.

@cursor_ai Please let us hide the primary bar 😶 who does that to a developer tool

@cursor_ai make it more reliable

@cursor_ai Can I load agents with .md for work instruction and role training before I start with an agent? Right now I have to do a couple prompt trainings before getting a decent agent that isn’t gonna “hallucinate” or make junk code.

@cursor_ai The @taggng of files - especially new files seems to be indexing much quicker. Maybe it’s just me 🤷

@cursor_ai Curious, is composer-1 a bit slower than the others? for instance, i recently compared it with windsurf's swe-1.5, and the difference was significantly high! Here's the comparison test I did! https://youtu.be/aFQgUK3pgoA?s...

@cursor_ai Cursor knows claude ide is on the way which can potentially cause the inevitable end of this product.

@cursor_ai Please God fix the bug on Mac OS Tahoe where it randomly just stops copying text I'm trying to copy.

@cursor_ai Do you have any weekly usage limits? If yes where to track it?

@cursor_ai Cursor using cursor to improve cursor 🙂

@cursor_ai Why did you remove the close button of the chat?

@cursor_ai Done with the dumb cafes and going back to focussing on product?

@cursor_ai every time i open cursor you want me to install a new update

@cursor_ai Gj boys

@cursor_ai Awesome 👏

@cursor_ai Kewl

@cursor_ai ❤️

@cursor_ai Incredible stuff

@cursor_ai so good

@cursor_ai Great

@cursor_ai Ship the student promotion next.

@cursor_ai Курсор, Кодекс и Копилот - это костыли для тех, кто пишет по инструкции, не чувствуя живой логики систем. Они помогают воспроизводить старые паттерны, но не создают новые формы.