HTML parsing has more edge cases than it looks like it should — unclosed
tags, implicit closing rules, mismatched nesting — and a hand-written
parser is exactly the kind of thing that gets those subtly wrong. Our
new HTML to Markdown Converter
sidesteps that entirely: it hands your HTML to new DOMParser(), the
same engine your browser uses to render any page you visit, then walks
the resulting, already-correct tree.
- Real parsing, not regular expressions over raw text — every edge case the HTML spec already defines correct behavior for is handled for free, because the browser’s parser is doing the actual parsing.
- The everyday subset, done honestly — headings, bold/italic, links, images, single-level lists, blockquotes, horizontal rules, and both inline code and fenced code blocks. Nested lists and tables are outside what this tool attempts, and the FAQ says so directly.
- Whitespace handled the way a browser renders it — ordinary text
has its whitespace collapsed like normal HTML rendering, while
<pre>and code blocks keep their exact original content. - Private by default — parsing and the tree walk both happen in your browser; nothing you paste is ever sent to a server.
Paste in a snippet of formatted HTML — from a CMS export, an email, or a page you copied — and get back clean, readable Markdown instantly.