I built a GraphQL API to cache the responses and dataload them like 2 years ago on top of HN API so it can give nicer responses and don't hit the endpoint as mu

I built a GraphQL API to cache the responses and dataload them like 2 years ago on top of HN API so it can give nicer responses and don't hit the endpoint as much through batching. It can load entire HN thread in one request. https://github.com/searchableguy/hackernews_graphql https://github.com/searchableguy/hackernews_graphql It is still incomplete though because I remember you said I could not scrape the site beyond robot.txt limits which would be too less for the use case because I wanted to provide write functionality so I gave up. Maybe people can host a caching relay and a layer like graphql on top as an unofficial way to lessen the burden on the firebase API (it's a pain to use anyway).