Hey there!
Been a heavy user of Cursor AI on my evening and weekend side projects. Recently I’ve been making a push at my Fintech to use Cursor and switch from VSCode + Copilot’s new agent. So far the firm loves the idea of using Cursor but there is one hiccup, Cursor doesn’t verify that suggestions do not contain licensed code. This is a huge blocker for my firm to adopt this as it will 100% open us up to litigation. This has been a huge selling point of Copilot as it does this and does it very well.
I’m determined to find a way around this, so would love to see what this community thinks, or maybe someone from Cursor can chime in? One solution I’ve pitched but would require a more controlled rollout of Cursor is to do these checks at the pipeline level. This would allow us to lean on tools like Snyk to check for licensed code and block PRs from getting in. The only problem there is that the developer won’t know and will waste time only to find out he can’t use his solution.
Thanks!