This is really cool.

I did something similar with gemini cli by just wrapping it in tmux and building some extensions.[0]

Eventually that wasnt enough so I ended up forking it and adding REST endpoints to inject commands and read the screen.[1]

Your solution is much cleaner! I'll probably replace mine with it. Thanks for sharing!

[0] https://github.com/stevenAthompson/self-command

[1] https://github.com/stevenAthompson/gemini-cli-remote-control

I wrote a version of this post awhile back that gets into a bit more detail as to HOW to bolt on the determinism.

I'm glad to see others talking about it. One day we'll look back on this era the same way folks look back at the time before we validated inputs.

https://www.stevenathompson.com/effective-vibe-coding-best-p...

I hadn't heard of it, but it looks very cool. I'll check it out, thanks!

Bitwarden has been my favorite for some time. I was surprised to see them fair so poorly.

I actually expected the models to perform better than they did, but I'm a bit of an optimist. It makes sense for the more mature SAST tool to outperform them just based on maturity alone.

I love the idea of fighting the asymmetric hiring process with more tools for the candidate. I'm not sure how recruiters will feel though.

I wonder what percentage will just "nope" out of interacting with it?

Fair enough. To expand on that: you’re right, there is a friction hurdle. But we’re betting that the time-to-truth is more valuable to a recruiter than the 60 seconds it takes to chat with a Rep. If the Rep can verify a candidate’s salary requirements and technical stack in 2 minutes, it saves the recruiter a 30-minute wasted phone screen. That’s the ROI we’re aiming for.

https://www.stevenathompson.com/blog - I talk about some of my LLM experiments and plan to touch on some security related projects soon.

It was probably due to the lack of flow control. Serial doesn't work well when it's one-way. I did something similar to send logs waaaay back in the day, and it would constantly flip bits or send characters out of sequence, etc. I had to transmit very slowly to get it to work stably without any flow control. I want to say that I limited it to 9600 kbps before it started to become reliable.