Nice.
There have been some similar activities on this front. In case you didn't already know of them;
Literate Programming in the Age of LLMs - https://news.ycombinator.com/item?id=48691666
Natural Language Outlines for Code: Literate Programming in the LLM Era (inspired by LP) - https://dl.acm.org/doi/10.1145/3696630.3728541
Visual Programming + Literate Programming: A Better Way to Build Business Logic With AI - https://www.leapter.com/ai-verification-problem/
Renaissance of Literate Programming in the Era of LLMs: Enhancing LLM-Based Code Generation in Large-Scale Projects - https://arxiv.org/abs/2502.17441
This is wonderful, thanks! I was unaware of these!
You might also find my previous comment chains useful - https://news.ycombinator.com/item?id=48431591 https://news.ycombinator.com/item?id=48618662 and https://news.ycombinator.com/item?id=48264680
With AI agents we need some way to specify requirements, trace that they are met through the generated code and verify them against the requirements. All with the ability to generate documentation and facilitating understanding. Changes made to the specification should update code and vice-versa auto-magically. That is the holy grail of "Round-Trip Engineering".