HTML to Markdown

How this reads your output

Script and style blocks are removed first, so a page copied from a browser does not bring its behaviour along. Block elements are then converted in order of specificity, with code blocks handled before inline code so a fenced block is not mistaken for a run of inline spans. Entity references are decoded at the end, which is why the output reads as text rather than as markup.

What the results mean

Tables found

Table markup that could not be converted. Cells come through as plain text, so any table needs rebuilding by hand.

Nested lists found

Lists inside list items. They are lifted to a single level, so check the structure if the original nesting carried meaning.

Links

Anchors converted to Markdown link syntax with the href preserved. Link text is stripped of any inner markup.

Common problems and fixes

The output is full of stray blank lines and fragments

The source is probably a full page rather than a content region. Navigation, footers and layout wrappers all become empty lines once their tags are stripped. Copy just the article or main element from the browser inspector, which gives a much cleaner result.

Formatting inside links disappeared

Link text is flattened to plain text, since Markdown link labels cannot reliably carry nested emphasis across renderers. If the emphasis mattered, add it back by hand around the link rather than inside it.

Frequently asked questions

More in DevTools

• Markdown to HTMLRender Markdown to clean HTML, with code blocks escaped.
• HTML Entity Encoder / DecoderEncode or decode HTML entities, whichever the text needs.