I've been using similar workarounds to use ChatGPT programmatically and it's just not worth it. In my experience you just get blocked after a couple of requests. I'm really hoping they release an API soon.
I'm with you most of the time... but there are some commands that refuse to stick in my brain no matter how hard I try, and I get tired of looking them up
I implemented it as a runtime class in Python just to see, seems like you can get a lot of (this benefit) in other languages by doing the same: https://github.com/yasyf/python-secret-type
Really excellent point. The way we try to handle this is by sealing off the part of the system that has any access to customer credentials, and making sure those credential are as minimal as possible (e.g. we can grant and revoke permissions to an existing user, but we cannot create new users or new permissions).
We build primitives for best-practice security controls so you don't have to :) Basically, it's a bunch of building blocks that you can string together to help roll out workflows that ensure compliance with infosec policies.
Hey everyone! Yasyf, CEO of Sym here. We're so inspired by the HN community, and supporting builders and engineers in making their companies more secure is a big part of why we're building Sym.
If you have any questions or suggestions, please let me know! You can also check out a (slightly outdated) demo video of our SDK here: https://youtu.be/1r6pQDHRJo8
