Website to Markdown converter

Paste a link and get the page as clean Markdown: the main article with its headings, lists, tables, code blocks, images and links, without the menus, ads, cookie banners and footers. The page loads in a real Chromium browser first, so sites built with React, Vue or other JavaScript work too. Copy it or download a .md file.

Free to try, no signup. 5 conversions/day per visitor.

What you get

Most of a web page is not the content: the HTML of the Wikipedia article on origami is 433 KB, and its Markdown here is 68 KB. For the GitHub page of a library the difference was 353 KB against 7 KB, and 247 KB against 49 KB for an MDN reference page (measured on 22 October 2026). What is left is the text you came for, in a format that pastes cleanly anywhere:

Main article only by default, found with Mozilla's Readability, the library behind Firefox Reader View. Choose Whole page for home pages, lists and other pages without a single article.
Real structure: headings, bold and italic, ordered and nested lists, quotes, GitHub-style tables, fenced code blocks, images and links.
Absolute links, so every link and image still works after you paste the Markdown somewhere else.
Metadata: title, author, publish date, site name, language and word count, returned as separate fields by the API.

Why a real browser matters

Many sites send an almost empty HTML file and build the page with JavaScript. A converter that only downloads the HTML gets a loading spinner or nothing at all. This tool opens the URL in headless Chromium, waits for the page to load (or for the network to go quiet, if you choose Network idle), hides cookie banners if you ask, and only then reads the page.

Web pages to Markdown from code or for AI agents

The /v1/content endpoint of the Shotmint API does the same conversion. It takes the same options as screenshots (wait_until, delay, block_ads, hide_selectors, headers and cookies for pages behind a login), plus format (markdown, text or html) and main_content:

curl "https://shotmint.creasework.com/v1/content?url=https://en.wikipedia.org/wiki/Origami" -H "X-API-Key: YOUR_KEY"

It answers with JSON: url, title, description, byline, site_name, published_time, lang, main_content (false when no article was found and the whole page was used), format, words, content, truncated (content is capped at 1,000,000 characters), and status + final_url (the page's own HTTP status and final URL, so a pipeline can skip 404s). Feed it to a RAG pipeline, a summarizer or an n8n or Make scenario. AI agents can call it as the read_page tool of the Shotmint MCP server in Claude, Cursor or VS Code.

Each call is 1 credit, the same as a screenshot or a PDF. €5 buys 2,000 with no subscription, and credits never expire.

Get 25 free credits a month Read the API docs

FAQ

Why is part of the page missing? In Main article mode, sidebars, comments and related links are dropped on purpose. Switch to Whole page to keep everything visible. If text appears only after a click or a scroll, it is not in the page when it loads, so it can't be extracted.

Does it work on pages that need a login? Not in this free tool. With the API, send your session cookies or an authorization header in a POST body; they only go to the site you convert.

Can it convert a whole site? One page per call. Loop over your sitemap URLs with the API to convert a site.

What about PDFs? To save a page as a PDF instead, use URL to PDF. For text inside a PDF file, use PDF to text.

Website to Markdown converter

What you get

Why a real browser matters

Web pages to Markdown from code or for AI agents

FAQ

Related tools