[hendler]

I also wrote wrote an LLM to Prolog interpreter for a hackathon called "Logical". With a few hours effort I'm sure it could be improved.

https://github.com/Hendler/logical

I think while LLMs may approach completeness here, it's good to have an interpretable system to audit/verify and reproduce results.

[shchegrikovich]

This is really cool!

[hendler]

Thanks! Feel free to reach out.