Markdown to HTML Converter

Turning Markdown source — the plain-text syntax with # for headings and ** for bold — into the HTML a browser actually renders. A parser reads the Markdown, works out the structure, and emits the equivalent elements: # Hello becomes <h1>Hello</h1>.

Paste or type your Markdown on the left. The HTML appears immediately on the right — there is no Convert button because there is nothing to wait for. Switch to the Preview tab to see it rendered, then copy or download either the HTML or your original Markdown.

No. Markdown is a source syntax designed to be readable as plain text; HTML is the markup browsers render. Markdown is deliberately limited — it covers the structures most writing needs and nothing else. HTML can express anything, at the cost of being far more verbose to write by hand.

Yes, six levels, using one to six # characters. # Hello becomes <h1>Hello</h1> and ###### becomes <h6>. The underline styles — text followed by === or --- — also work for the first two levels.

Not in the original spec, but yes in GitHub-Flavored Markdown, which this converter enables by default. Pipe-delimited rows with a --- separator become a semantic <table> with <thead> and <tbody>. Column alignment set with :--- and ---: is preserved as an align attribute.

Yes. Fence a block with triple backticks and it becomes <pre><code>. A language after the opening fence becomes a class — ```javascript produces <code class="language-javascript">, which is the convention every syntax highlighter reads. Indented four-space blocks work too. Content inside a code block is escaped, so markup in your example code stays visible as text.

Yes:  becomes an <img> with the alt text preserved as the alt attribute — which matters, because that is what a screen reader announces. This converter does not upload, proxy or fetch images; the URL is passed through as you wrote it.

Yes: [text](https://example.com), plus autolinks like <https://example.com>. External links get rel="noopener noreferrer" added. No target attribute is set — whether a link opens in a new tab is a decision for the page the HTML ends up in, not for a converter.

A small, predictable set: h1–h6, p, strong, em, del, code, pre, blockquote, ul, ol, li, a, img, hr, br, the table elements, and a disabled input for task list checkboxes. That is the whole vocabulary — which is what makes it practical to guarantee the output is safe.

Markdown permits it, and most parsers pass it straight through. This converter does not: raw HTML in your source is escaped and appears as visible text, so <div> comes out as &lt;div&gt;. That is a real limitation and it is deliberate — see the next answer.

Because passing it through safely requires a full HTML sanitizer, and half-doing it is worse than not offering it. If raw HTML flowed through, so would <script>, event handler attributes and everything else in an untrusted document. Escaping it is a clear rule with no gaps: every character of your Markdown either becomes markup this converter generated, or becomes text. If you need raw HTML preserved, this is not the right tool for that document.

The HTML this converter produces is, because of how it is produced rather than because of what is filtered afterwards. Raw HTML is escaped, and link and image URLs are checked against a four-scheme allowlist — http, https, mailto and tel. Nothing else can appear. The preview additionally re-checks the output against an element and attribute allowlist before rendering it, using the browser's own parser.

Because a Markdown parser is not a security boundary and does not claim to be. The parser used here removed its sanitize option in version 8 and points users at a dedicated sanitizer instead. Feed it <script>alert(1)</script> and it emits exactly that. Any application rendering parser output from untrusted Markdown — comments, user profiles, wiki pages — without sanitizing has an XSS.

Parsing decides what the Markdown means and produces markup. Sanitization decides what markup is allowed to survive. They are separate jobs and a parser doing the first well tells you nothing about the second. This tool solves it by never generating dangerous markup in the first place, then checking the result anyway before it renders.

Only via HTML — a <script> tag, an event handler attribute like onerror, or a javascript: URL in a link. Markdown itself has no syntax for code execution. This converter neutralises all three: script tags are escaped as text, event handlers go with them, and javascript: URLs are refused while the link text is kept.

Yes, along with vbscript:, file: and anything else outside the allowlist. Obfuscated forms are caught too — a tab or newline inside the scheme, mixed case, or spaces before the colon — because browsers strip those characters before deciding what scheme a URL has, and a naive check that does not is exactly how these get through. data: is blocked even for images, because data:image/svg+xml can carry script.

Yes, with Copy HTML. Copy Markdown copies your original source instead. Both use the browser's clipboard API; nothing is sent anywhere.

Yes — Download HTML saves document.html and Download Markdown saves document.md. Both are generated in your browser with a Blob and never uploaded. With the full-document toggle on, the HTML download is a complete page you can open directly; with it off, it is the fragment.

It wraps the converted fragment in a minimal valid page — doctype, html with a lang attribute, a UTF-8 charset declaration, a viewport tag and a title. The default is off, because what people usually want is the fragment to paste into a template or a CMS field. No CSS is included: a converter picking your typography would be the first thing you deleted.

Open the file, select all, paste it in. There is no file picker because the paste path covers it and adding one would mean handling encodings and file sizes for no real gain. Large documents are fine — a 60 KB file converts in a few milliseconds.

No. Parsing happens entirely in your browser using a library bundled with the page — there is no backend, no request, and nothing stored. Unlike some tools here, this one also deliberately does not put your content in a shareable URL: Markdown documents are long and often private, and a link would put them in browser history.

Yes. Conversion, preview, copy and download are all free, with no account and no limit.

Yes — that is what the fragment is for. Paste it into a template, a component, or a static site's HTML. It carries no classes beyond the code block language and no inline styles, so your own CSS decides how it looks.

Usually. Most CMS rich-text and raw-HTML fields accept it. Some strip elements they do not expect, some rewrite attributes, and email clients are stricter still — so test it in the destination rather than assuming. No converter can promise output that survives every CMS and email client unchanged.

Because HTML describes structure and CSS decides appearance. The same <h1> is 32 pixels on one site and 20 on another; a <blockquote> may have a coloured bar or nothing at all. The preview here uses this site's typography so the structure is legible — it is not a prediction of how it will look anywhere else.

GitHub's superset of the original spec, and the dialect most people mean by "Markdown" now. It adds tables, strikethrough with ~~text~~, task lists, autolinking of bare URLs, and fenced code blocks with language identifiers. This converter enables it by default.

Yes — - [x] done and - [ ] todo become checkboxes. They are rendered with the disabled attribute on purpose: a preview is for reading, and a checkbox you can toggle implies a state the document does not have.

Formatted, with block-level elements indented by nesting depth. The indentation is added only to lines that already begin with a block-level tag, and code block contents are left untouched — reindenting a code block would change the code inside it. What you copy is exactly what you see.