Generally you'd just generate the .md version, not create different docs.
But, there's no reason why they can't parse HTML, assuming it is well written semantic HTML, not a 1000 nested <div>s ... and there is no real proof I've seen that having llms.txt is helpful.
Generally you'd just generate the .md version, not create different docs.
But, there's no reason why they can't parse HTML, assuming it is well written semantic HTML, not a 1000 nested <div>s ... and there is no real proof I've seen that having llms.txt is helpful.