URL to Markdown

Instructions

1. Enter a URL — paste the web address of the page you want to convert into Markdown.
2. Choose what to include (optional) — every toggle starts off. Flip on the document title, meta description, nav, header, footer, aside, or JSON-LD structured data if you want any of them kept.
3. Convert — click "Convert to Markdown". The result lands in the output box, and "Copy" puts it on your clipboard.

Using the URL to Markdown converter

I built this because I kept trying to paste entire pages into a ChatGPT conversation, and the HTML source was always too large to fit in a single message. Pruning the HTML down to the meaningful parts by hand was a headache. Markdown gives you all the important content without any of the HTML noise.

The page is fetched server-side, parsed, and converted — so what you get back is the Markdown of the HTML the server delivered, before any JavaScript ran. Only public URLs work; there is no access to your cookies or session, so pages behind a login come back as whatever a logged-out visitor sees.

What the converter actually does

Because the toggles all default to off, a plain conversion gives you the body content only — no title, no nav, no footer, no sidebar. That's usually what you want when the destination is a text box with a size limit.

Convert a website to Markdown for LLMs

This one's my favorite, and it's why the tool exists. Feeding a model raw HTML wastes context on markup that carries no meaning: class names, inline styles, tracking scripts, cookie banners, nav menus repeated on every page. Converting to Markdown first:

• Gives the model structured, noise-free input, which tends to produce better comprehension and more relevant output.
• Keeps only the sections of the page you actually care about.
• Cuts preprocessing time — you're not hand-deleting ads, menus, and scripts before you paste.

If you're assembling context for a model, leave the toggles off and take the body. If you're building a retrieval corpus where provenance matters, turn on the title and meta description so each document carries its own header, and turn on JSON-LD when the page publishes structured data worth keeping (article metadata, product fields, FAQ entries).

Content repurposing

Converting a webpage into Markdown gives you a lightweight, portable format that drops straight into most text editors and CMS platforms. It's a fast way to pull web content into blogging, documentation, or knowledge-sharing workflows without carrying a page's styling along with it.

Markdown for GitHub

Markdown is the native format of GitHub, so this is a quick way to extract and format documentation, tutorials, or technical content from the web and drop it into a repository. Its simplicity is what makes it good for collaboration and version control — a Markdown diff is readable, an HTML diff usually isn't.

FAQ

How do I convert a URL to MD?

Paste the URL above and click "Convert to Markdown". The page is fetched, stripped of scripts and (by default) its nav, header, footer, and sidebar, then converted. Copy the result out of the output box.

Can I convert a website to Markdown for an LLM?

Yes — that's the main use case. Leave every toggle off to get just the body content, which is the densest, lowest-noise version of the page. Add the title and meta description back if you need each chunk to identify itself.

Does it work on pages that need a login?

No. The fetch is anonymous and server-side, so a page behind a login returns whatever a logged-out visitor gets — usually a sign-in page.

Why is the output nearly empty for some pages?

Because the conversion runs on the HTML the server returned, before any JavaScript executes. A page built entirely client-side ships almost no content in its initial HTML, so there's little for the converter to work with. The same limitation applies to anything reading the raw source rather than the live DOM.

What happens to links in the converted Markdown?

Relative links are made absolute against the page's origin before conversion, so a link written as /about becomes a full URL. They keep working after you paste the Markdown somewhere else.

Can I keep the structured data?

Yes. Toggle on "JSON-LD structured data" and every application/ld+json block on the page gets appended to the end of the output under a ## JSON-LD Scripts heading, each one in its own fenced JSON block. Scripts are otherwise always removed.