I’m a paying for the $20 subscription and I love Cursor for quick prototyping (some call it Vibe Coding). However, I am frequently hitting my 500/500 request limit with the $20 subscription plan. I know I can pay for extra requests and put a spending cap; but I’m wondering if there are other things I can do. I feel like some of the responses from the AI model are not great and I constantly have to ask it questions in another way in order to get a good result with good code. This eats up additional requests. Are there things I can do to get more efficient with my prompts so I do not need to go through a longer trial/error cycle eating up requests.
The best you can do is to dedicate time to write good project rules, including the description of your project - goals, code structure, layout, styleguide, and everything that might be important for the model to code the way you want it.
Tell it exactly how you expect it to behave (each of us has different preference of how to work with Cursor), and use models that work the best with your particular set of requirements.
Tell it to document (in a file, so there’s a track of it) the tasks it plans or suggests to perform, while focusing on performing the ones you ordered it to.
Stack those tasks - some models work well performing multiple actions over one response, as long as they’re adequately prompted.
Delegate easier tasks or prompts to free models.
I pretty much consider all of this a requirement for a Cursor user who wishes to save fast requests.