Fetch

One endpoint per domain, with the extraction config already resolved. Spider works out the configuration for a site on the first request, validates it, and caches it. Every fetch after that reuses the cached config, so extraction is faster and more consistent. It amounts to a custom scraper per website, built for you.