How This Site Is Built

A short colophon: how vodfetch is put together, and why it is built to be read by both people and machines.

Why it reads like this

More and more of the people 'reading' a website are machines — search crawlers, answer engines and AI assistants. vodfetch treats them as a first-class audience rather than an afterthought, without ever hiding anything from humans: the exact same content is served to both.

This isn't a growth hack bolted on afterwards; it's how the site is built from the ground up.

Plain by design

Every page is server-rendered static HTML. There is no login wall, no cookie gate, and no JavaScript required to read the content — the downloader itself uses JavaScript, but the words, facts and links all live in the HTML.

We use system fonts only, so there are zero external font requests and nothing render-blocking between you and the page.

Two formats for every page

Append ".md" to any page URL to get a clean Markdown version of it. Machine-readable summaries of the whole site live at /llms.txt and /llms-full.txt (per language), structured facts at /facts.json, and a canonical entity description at /grounding.json following the Grounding Page Standard.

Nothing there is hidden or cloaked: the Markdown and JSON simply restate, in a machine-friendly shape, exactly what the HTML says.

Structured for meaning

Every HTML page carries a single JSON-LD @graph — SoftwareApplication, FAQPage, HowTo, BlogPosting, BreadcrumbList — with stable @ids, so the relationships between entities and pages are explicit rather than guessed.

A full XML sitemap lists every page, and hreflang tags connect all fourteen language versions of each one.

Open and verifiable

vodfetch is open source under the MIT license on GitHub, so the tool's behaviour is auditable rather than a black box. The editorial standards behind the words are written down too — see the Editorial & Honesty Policy.

The short version: human-readable and machine-readable are the same thing here — plain HTML, no gates, every page mirrored as Markdown and described in JSON, all of it crawlable and checkable.