Feature request for product/service
AI Models
Describe the request
Is it possible to bake security checks into Composer model itself, or its tooling? The more accessible “programming” gets, the more inexperienced programmers are producing and releasing applications, where no effort is put into security. Which makes sense - if I have little experience in programming, how would I know what I don’t know?
Is it possible to bake this into the tools or models somehow? I understand that for simple POC projects I might want to turn this off, but as soon as we are talking about production code, on enterprise level, publicly launched apps which exchange and store sensitive user data, there should be a separate background agent that is constantly surverying the entire codebase for vunerabilities, and checking itself. Even as an experienced programmer, at the speed which I am able to move with Cursor, my main bottleneck is verifying security. Granted, I can create my own agent to do this for me, but inexperienced users would not be aware of the need to do that. Many vunerabilities take time to emerge in practice. Therefore it would be great if a security agent would come with Cursor out of the box, and would be turned on by defult.