I've run into these problems when building with AI: - AI loses context - need a bunch of project rules, cursor rules, system prompts but the AI still doesn't follow them sometimes. - AI can't fix bugs - telling it "that didn't work" 100 times doesn't work - stuck at 60% - hard to make progress once project gets more complex, it add one feature but breaks 3 other. So i've decided to build a cursor clone (which is really a vs code clone) that has - better context management - AI maintains context of project scope, feature requirements, task breakdown, and testing. - visual debugging - AI traces code execution and isolates issues much better (equivalent of putting console logs everywhere, but you can see where the bug is) - native integration with supabase - a lot of the issues I run into are database related, so this should help. It's working very well with sonnet 3.7 & deep-seek, but it could just be my own test cases. I'm looking to have a few people test this. Also, should I open source this?? If you are interested, leave a comment :)

cursor.directory

Has anyone built/found either a cursor or windsurf rules file to optimize for building crews?

Is anyone here targeting people/businesses that use human Virtual Assistants/VAs and attempting to replace them with some form of AI Agents? If so, how is that going? Have you been successful at replacing the repetitive tasks of a low paid human, and if so, how much is this worth to your clients as a percentage of what they were previously paying their VAs?

After exporting a project from bolt.new, where to continue?