Turn any page into Markdown
Paste HTML or a URL and get the closest Markdown equivalent: headings, lists, tables, code blocks, links and images all map to GitHub-flavored Markdown. Scripts, styles and page furniture are dropped, nothing is rewritten or summarized.
The conversion is mechanical. Each HTML element maps to its closest Markdown construct, and elements with no Markdown equivalent (scripts, styles, form fields, SVG) are dropped, never paraphrased. Relative links and images are resolved to full URLs so they keep working outside the page.
Pasted HTML is converted entirely in your browser and never sent anywhere. In URL mode our server fetches the page's HTML for you (browsers cannot fetch other sites directly) and the conversion still happens in your browser. Nothing is stored.
Pages built by JavaScript can look empty here: this tool reads the HTML the server sends, the same thing most crawlers and Markdown pipelines see.
Moving content, not just one page?
BeeRanked's free content importer reads a whole blog or content hub, article by article, and hands you everything as structured data. And the platform publishes it back on your own domain with the SEO work built in.
Start with BeeRankedOr try the free content importer on your whole site.
What this converter does
Paste HTML, or a URL, and the page comes back as GitHub-flavored Markdown. Headings, paragraphs, lists, tables, code blocks, links and images each map to their closest Markdown form; scripts, styles, form controls and page furniture (navigation, header, footer, sidebars) are dropped. Nothing is rewritten and nothing is summarised: what has no Markdown equivalent is left out, never paraphrased.
The conversion runs in your browser. In URL mode the server fetches the page's HTML and hands the bytes back unchanged, because a browser cannot fetch another site on its own; pasted HTML never leaves your machine. When a page has a <main> or <article> element the converter starts there, so the output is the content, without the chrome around it.
Why GitHub's flavour
Markdown has several dialects. The converter writes GitHub-flavored Markdown (GFM), which its specification defines as "a strict superset of CommonMark": everything CommonMark renders, plus tables, strikethrough and task lists. It renders on GitHub, in documentation tools and in every Markdown editor, and it is the format BeeRanked's importer and editor read.
The output is also CommonMark-safe: a literal asterisk, or a line that starts with "1." in the source text, is escaped, so rendering the Markdown back cannot change the structure of the document.
Markdown is the format machines ask for now
Two documents explain why a Markdown copy of a page is worth having. Cloudflare's Markdown for Agents (12 February 2026) measured its own announcement: "This blog post you're reading takes 16,180 tokens in HTML and 3,150 tokens when converted to markdown." Sites on Cloudflare can now serve that version on request: "the client needs to add the Accept negotiation header with text/markdown as one of the options."
The llms.txt proposal asks for the same thing at the site level: "We propose adding a /llms.txt markdown file to websites to provide LLM-friendly content", a file that "offers brief background information, guidance, and links to detailed markdown files". A page converted here is one of those detailed files.
Where the conversion is lossy
- Merged or nested table cells. Markdown tables are plain grids, so the cell text is kept and the merging is lost.
- Embedded frames, video and audio. Markdown has no syntax for them, so they are reduced to a link or dropped.
- Text styling beyond bold, italic and code. Colours, underlines and font sizes have no Markdown and disappear.
- Pages built by JavaScript. URL mode reads the HTML as the server sends it; if the content is added by scripts after load, the Markdown will be short. Paste the rendered HTML from your browser's inspector instead.
- Very long pages. The URL fetch stops at 1.5 MB or 8 seconds.
Three things to do with the output
- Move a page. Markdown is the input format of static site generators and of BeeRanked's editor and MCP tools. Convert, check the headings, import.
- Brief an assistant. A Markdown copy carries the page's structure at a fraction of the tokens, so an AI assistant reads the whole page instead of a truncated one.
- Check the structure. Without the layout, the heading order, the missing alt texts and the link texts are easy to see, which is close to the view a crawler has.
Questions people ask
- Is the page stored? It is not. URL mode fetches and returns the HTML; the conversion happens in your browser and nothing is written on our side.
- Why is the result empty or short? The page is built by JavaScript, or it blocks automated fetches. Open it in your browser, copy the HTML from the inspector and use the paste tab.
- Can it convert a whole site? One page at a time. For a blog or documentation hub, the content importer finds every article and returns them all as one file.
- Is it free? It is, with no sign-up and no email.
Read more
- llms.txt in our wiki: what the file is and what goes in it.
- llms.txt: what it is, and whether your site needs one.
- Controlling what AI crawlers see: which agents read your pages, and the rules for each.