Word to HTML / Markdown / Text
Turn a Word document into clean HTML, Markdown or text — style by style.
Output
Styles in this document
Choose what each Word style becomes. “Default” keeps the converter’s own choice.
Result
About the Word to HTML / Markdown / Text
Turn a Word document (.docx) into clean, semantic HTML, Markdown or plain text — and decide what each of its styles becomes. The tool lists every paragraph and character style the document actually uses, how often and with an example, so a custom “Section Title” can become a level-2 heading, “Source Code” a code block and “Pull Quote” a block quote, while a style you don’t want, such as an internal note, is left out. Headings, lists, tables, links, bold and italic, footnotes and endnotes are kept, and comments too if you want them; fonts, colours and Office styling are not.
Get HTML for a CMS or an e-mail template (a fragment or a complete page, with heading anchors and a table of contents if you like), Markdown for GitHub, a wiki or a static site, with tables and footnotes, or plain text with aligned tables and optional line wrapping for e-mail. Pictures can be embedded, saved as separate files in a ZIP, or left out. The file is checked, converted and previewed in your browser and never uploaded.
How to use it
- Choose a .docx file — saved from Word, LibreOffice, or Google Docs with File › Download › Microsoft Word.
- Pick HTML, Markdown or Plain text, and what to do with pictures and footnotes.
- Under Styles in this document, choose what each style becomes, for example Section Title → Heading 2. Styles the converter doesn’t know are marked; Title, Quote and Code styles start with a sensible choice.
- Check the Preview, then Copy the result or Download it — as a ZIP when pictures are saved as separate files.
Examples
Paragraphs in the style “Section Title” (big and bold in Word, but not real headings)
Section Title → Heading 2: <h2>Overview</h2> (HTML) ## Overview (Markdown)
Revenue grew 12%¹ … ¹ Audited figures.
Revenue grew 12%[^1] … [^1]: Audited figures.
A Word table with the rows Region | Sales and North | 1,200
Region Sales ------ ----- North 1,200
Two lines in the style “Source Code”: SELECT * / FROM sales;
``` SELECT * FROM sales; ```
Common uses
- Publishing articles, policies or documentation written in Word on a website or CMS without Word’s hidden formatting.
- Moving a Word manual or report to Markdown for GitHub, a wiki or a static site generator, footnotes and tables included.
- Pasting a document into a plain-text e-mail or a system that takes no formatting, with tables that still line up.
- Getting every picture out of a document as separate files next to the HTML or Markdown.
Styles in this document
Word documents usually mark structure with styles: Heading 1, Quote, or a company’s own Section Title. The converter (mammoth) knows Word’s built-in heading styles; any other style becomes an ordinary paragraph, which is why custom headings often arrive as plain text. The list shows each style the document uses with how often it appears and the start of its first paragraph, and you choose what it becomes: a heading level, paragraph, block quote, code block, list item, or left out. Character styles (for words inside a paragraph) can become bold, italic, code, highlighted, underlined or struck-through text, plain text, or be left out.
List Paragraph is the style Word gives most numbered and bulleted items. Leave it at Default — the lists come from Word’s numbering — unless you really want those paragraphs turned into something else.
What each output keeps
- HTML: headings, paragraphs, line breaks, bulleted and numbered lists (nested too), tables with merged cells, links, bold, italic, strikethrough, superscript, subscript, code, block quotes, pictures with their alternative text, and footnotes, endnotes and comments as numbered lists at the end, linked both ways. No classes, styles or fonts. A complete page gets the document’s language (
<html lang>) when the document states one. - Markdown: CommonMark with GitHub’s additions — pipe tables (the first row becomes the header), strikethrough written
~~text~~and footnotes written[^1], which GitHub and many other renderers show as linked notes. Tables with merged cells or lists inside cells, and superscript, subscript and underlining, stay as HTML, which Markdown allows. - Plain text: headings underlined with = and -, lists with - and 1., tables as aligned columns, links as “text (address)”, pictures as [Image: alternative text] and notes at the end as [1]. Lines can be wrapped at 72 or 80 characters for e-mail; tables and code blocks keep their width.
Safety: the archive check and sanitising
A .docx file is a ZIP archive in the Office Open XML format (the ECMA-376 standard). Before converting, the tool reads the archive’s table of contents and refuses files with more than 10,000 parts, more than 600 MB when unpacked, encrypted parts, or parts (text or pictures) that unpack to more than they declare or are compressed far more than real documents are — the signs of a “zip bomb”. The conversion runs in a separate worker that is stopped if it takes longer than two minutes.
The resulting HTML is sanitised with DOMPurify against a list of document tags: scripts, event handlers, styles, frames, forms and javascript: links are removed. The preview only shows pictures stored inside the document, so opening a file never loads anything from the internet.
Limitations
- Old Word 97–2003 .doc files, password-protected documents, and .odt or .rtf files can’t be read: save them as .docx (without a password) first.
- Layout isn’t converted on purpose: fonts, colours, sizes, alignment, columns, headers and footers, page numbers and table borders are dropped. Text in text boxes appears as ordinary paragraphs.
- Equations, charts, SmartArt and drawn shapes are left out. Tracked changes come out as if accepted: inserted text is kept and deleted text is left out.
- Pictures in Windows Metafile formats (EMF/WMF) are kept in downloads, but browsers can’t show them, so the preview shows a labelled box instead.
- One document at a time; files over 200 MB are refused.
Privacy
Everything happens in your browser. What you enter or open here is not uploaded or stored by MySmartCoPilot.
Frequently asked questions
How is this different from the Word to HTML Converter?
Word to HTML Converter also cleans text pasted from Word or Google Docs, using a fixed set of rules. This tool works on .docx files and gives you more control over them: it lists the document’s own styles so you can decide what each becomes, writes plain text, turns footnotes into Markdown footnotes, and can add heading anchors and a table of contents.
Why did my headings come out as ordinary paragraphs?
They use a custom style (such as “Section Title” or “Chapter”) rather than Word’s Heading 1–6, and only the built-in heading styles are recognised automatically. Find the style under Styles in this document and choose the heading level it should be.
Can I get the pictures as separate files?
Yes. Under Pictures choose Separate files in a ZIP. The download is a ZIP with the HTML, Markdown or text file and an images folder (images/image-1.png …); the output refers to the pictures by those names, and identical pictures are stored once.
What happens to footnotes and comments?
Footnotes and endnotes are kept by default: in HTML as a numbered list at the end, linked both ways; in Markdown as [^1] notes; in plain text as [1] with the notes at the end. Comments are included only when you tick Include comments. Choose Leave out under Footnotes to drop notes and comments altogether.
Is the HTML safe to put on my website?
It is sanitised against a list of ordinary document tags — no scripts, event handlers, styles, frames or javascript: links — and contains no classes. Still read it through before publishing, as you would any document from someone else.
Is my document uploaded?
No. The file is read, checked and converted by your browser on your device. Nothing is sent to MySmartCoPilot or anyone else.