Benchmarks

On a 170-URL mixed benchmark run on 1 October 2026, 6 URLs no longer existed (both services got a 404); of the 164 live pages, Skryp returned 157 usable and Firecrawl 159.

What was run

170 public URLs in ten page types (articles, documentation, shops in South Africa and elsewhere, country-priced pages, listings, JavaScript apps, bot-protected pages, PDFs and our own websites), each read once by each service.

How it was judged

Usable means the result was the page itself (not an error, a block page or a 404) with at least 200 characters of text; completeness is the share of a neutral reference copy's main-content lines (a full browser render cleaned with trafilatura) found in the output.

What it does not show

One run from a South African connection before the move to a server; results from a data-centre connection may differ. Skryp's seven misses on live pages: three documentation pages answered Skryp with HTTP 503 during the run, two pages behind a Cloudflare challenge, a shop home page where Skryp kept only a banner and the cookie notice, and one page neither service read. Firecrawl's five: four listing and social pages and the same page neither read.

Next run

Re-run after hosting moves and after major routing changes.

On the same 170 public pages, Firecrawl returned 159 usable pages and Skryp 157; median completeness was 1.00 for Firecrawl and 0.98 for Skryp; on documentation pages Firecrawl returned 23 of 23 usable and Skryp 20 of 23.

What was run

170 public pages across articles, docs, shops, listings, JavaScript apps, PDFs, protected and regional pages.

How it was judged

Usable means the requested content was present and complete enough to answer from, judged against a reference copy; both services run on the same URL list.

What it does not show

One run. Firecrawl at default settings. Skryp's own benchmark; publish the URL list and scoring code with it.

Next run

Re-run before publishing a comparison and after major routing changes.

On 22 country-priced pages, Skryp and Firecrawl each returned the expected country's currency on 21.

What was run

22 regional shop pages in the 170-page benchmark.

How it was judged

The dominant currency in the output was compared with the requested country.

What it does not show

Currency is a proxy for the right regional page, not proof that every price is correct.

Next run

Re-run with the benchmark.

Asked for 20 countries with strict country on, Skryp returned the page through an exit verified in the requested country in 19 of 20.

What was run

One public page per country, 20 countries, strict-country requests, residential and mobile routes.

How it was judged

Each request's exit IP was checked with a geolocation service asked through that exit (ipinfo.io first, then fallbacks; ISO codes only) before the page was read; the receipt records the result.

What it does not show

A single run of 20 pages. Exit verification proves where the request left from, not that every field on the page is correct for that market.

Next run

Re-run after routing or proxy-supplier changes, and at least monthly.