HTML to Markdown
Turn HTML into clean Markdown โ headings, links, lists, tables.
About this converter
Parses your HTML as a real DOM and walks it, emitting Markdown for everything Markdown can say: headings, bold and italic, links, images, nested lists, fenced code from <pre>, quotes, rules, pipe tables. Scripts, styles and nav clutter are dropped on purpose.
Rescuing content from markup
The job is almost always migration: a blog leaving its CMS for a static-site generator, documentation headed into a README, articles moving to a Markdown-first platform. What makes conversion good is knowing what NOT to keep โ Markdown's deliberate poverty (no colored spans, no layout divs, no font stacks) is its portability, so this converter degrades unmappable styling to plain text instead of emitting HTML islands that defeat the purpose. Structure that maps, maps faithfully: <pre> becomes a fenced code block byte-for-byte, nested lists keep their nesting, tables become pipe tables with their pipes escaped. The DOM-walk approach (rather than regex guessing) means malformed HTML gets the browser's own famously forgiving repair before conversion โ the same repair your readers' browsers were applying anyway. Workflow tip: converting a whole page? Paste just the article's HTML, not the full document โ less nav junk to discard. The round trip runs through the Markdown preview, tables from scratch come from the table generator, and word frequency audits the rescued prose.
Frequently asked questions
What happens to styling Markdown can't express?
It degrades to plain text deliberately โ colored spans and layout divs lose their decoration, keep their words. Emitting raw HTML islands would defeat the point of converting.
Are tables supported?
Yes โ HTML tables become Markdown pipe tables, with pipes inside cells escaped. Complex row/column spans flatten (Markdown tables have no spans), which is the format's honest limit.
Why did scripts and navigation disappear?
By design โ <script>, <style> and <nav> are dropped because content conversion wants the article, not the page chrome. Paste the article fragment for the cleanest result.
Does malformed HTML break the conversion?
No โ input parses through the browser's own error-tolerant HTML engine (the one that fixes tag soup on every page you visit), so conversion sees the repaired DOM.