How do I convert HTML to Markdown?

Paste your HTML into the HTML input box, or use Load example. The Markdown output updates as you type. Press Copy, or Download .md to save the result as a file.

Can I paste directly from a web page, Word or Google Docs?

Yes. When you paste rich text, the tool captures its HTML automatically, so you do not need to view the source first. The Markdown output appears on the right.

Which HTML elements are converted?

Headings, paragraphs, links, images, bold, italic, strikethrough, ordered and unordered lists, blockquotes, code blocks, horizontal rules and basic tables become Markdown. Tables become pipe tables with the first row as header.

Why did my styling, scripts or class attributes disappear?

Markdown has no equivalent for them, so inline styles, classes, scripts and head content are dropped. Complex tables are simplified, so check the output before publishing.

Is my HTML sent to a server for conversion?

No. The conversion runs in your browser with hand-written code and no third-party library, so your HTML stays on your device. It does not fetch images, so image links still point to the original host.

Yes. Every tool is free to use, with no account and no credit card. The work runs on your device rather than on our servers, which is what keeps it free to run.

Does it work on mobile?

Yes. It runs in current mobile browsers on iOS and Android, with nothing to install.

📝

Free HTML to Markdown Converter

Convert HTML to clean Markdown. Handles headings, lists, links, images, code blocks, tables, and blockquotes. Perfect for moving content out of CMSes.

Runs in your browserNo sign-up, no pop-upsNo login requiredFree forever

Works offline once loaded

Paste HTML on the left; clean Markdown appears on the right. Markdown is simpler than HTML, so styling, scripts, and complex tables are simplified or dropped.

HTML input

Tip: paste rich text copied from a web page, Word, or Google Docs - its HTML is captured automatically.

Markdown output

Markdown output will appear here…

Common uses

Migrate Blog Content to MarkdownConvert old WordPress, Medium, or other CMS posts to Markdown for static site generators.
Convert Email Content to MarkdownPull text from HTML emails or web articles and convert to plain Markdown for note-taking apps.
Convert HTML Docs to MarkdownMove legacy HTML documentation to a Markdown-based docs system (Docusaurus, GitBook, MkDocs).
CMS Export CleanupClean up the messy HTML exported from CMSes and turn it into version-control-friendly Markdown.
Prepare HTML Content for LLM PromptsConvert HTML content to clean Markdown for use in ChatGPT, Claude, or other LLM prompts.

How It Works

This converter is built around a hand-written, zero-dependency regex pipeline — no third-party library is imported. The entire conversion runs inside your browser via a React useMemo call, updating the output live as you type.

1
Paste detection intercepts rich text
A ClipboardEvent handler reads the text/html flavor from the clipboard before the browser can paste plain text. This means pasting directly from a web page, Google Docs, or Word automatically captures the raw HTML markup — you don't need to view source first.
2
Multi-pass regex pipeline strips and converts tags
The converter runs a strict ordering of regex passes: it first removes <head>, <script>, <style>, and HTML comments, then stashes fenced code blocks behind null-byte placeholders to prevent later passes from corrupting code content. Subsequent passes handle headings (h1–h6), blockquotes, images, links, bold/italic/strikethrough, ordered and unordered lists, and basic GFM pipe tables — each pass targets a specific tag family.
3
Entity decoding and whitespace normalization
After all tags are stripped, a dedicated decodeEntities function resolves named entities (&, <, etc.) plus numeric decimal and hex character references (', ') using String.fromCodePoint. Malformed or out-of-range code points are replaced with the Unicode replacement character rather than throwing. Trailing spaces, collapsed runs of blank lines, and excess whitespace are then normalized.
4
Protected code spans are restored
The placeholder system (null-byte-delimited indices) is unwound last, after stripping, decoding, and whitespace collapsing are complete. This prevents entity sequences inside code literals from being decoded twice — a common double-decode bug in naive HTML-to-Markdown converters.