Using an LLM with CAS math validation tools to validate it's work and then to generate a markdown document with CAS blocks:

  :::cas mode=chain engine=sympy assumptions="x \neq 1"
  $$ \frac{x^2 - 1}{x - 1} $$
  $$ = \frac{(x-1)(x+1)}{x-1} $$
  $$ = x + 1 $$
  :::

Which you can run to check it.

Thanks for the feedback, I originally tried to import all domains ~1b, but that took a week, so I'll always be a week behind unless I do some serious optimisation there! Currently using godaddy API, which has FAST or ACCURATE options. Given the generator requests are already 10s long (most of which is gpt4) I opted for the FAST, but I can trial it with the ACCURATE and see how it goes. Thanks again!

Maybe you could take a look at a Bloom Filter datastructure, space efficiënt probalistic set for testing if something is present

[1] https://en.m.wikipedia.org/wiki/Bloom_filter

Thanks for the feedback, I added a handful more TLD's, but need to think harder about the UX for the full list. The corporate feel is interesting, maybe adding a tone dimension like formal, neutral, light hearted could be interesting. You can also add the tone to the description although that's not obvious, I was thinking of adding some examples to help.