md to html without uploading your files: a browser-based converter that doesn't choke on Mermaid or KaTeX

Every few weeks I need to hand a Markdown document to someone who doesn’t live in a text editor. A README for a client, meeting notes, a short spec with a couple of diagrams. And every time, the same two problems showed up:

1. Anything beyond plain GFM breaks. The moment a document contains a Mermaid diagram or LaTeX math, a lot of the converters I tried give up. You get a code fence where a flowchart should be, or raw $...$ soup in the middle of a paragraph.
2. The documents I convert are the ones I least want to upload. Internal notes, drafts, client material. A lot of online converters process your files on their servers, and for work documents that always made me uncomfortable.

So I built a small tool that fixes both for me: a free md ↔ html converter that runs entirely in your browser. No install, no sign-up, and your files are never uploaded — conversion happens client-side.

What it does

• Both directions. Markdown → HTML and HTML → Markdown, on the same page. The HTML → Markdown side accepts a .html file, pasted HTML source, content copied straight from a web page, or a URL.
• GFM support: tables, task lists, strikethrough.
• Code blocks come out syntax-highlighted.
• Mermaid diagrams and KaTeX math actually render. The rendering engines are lazy-loaded only when your document uses them — a plain Markdown file never triggers those downloads at all.
• Batch conversion: drop several files at once, then download results individually or grab everything as one ZIP.
• Print / save as PDF with print-friendly styling, straight from the preview.
• UI is available in English and Korean.

The part I actually built it for: single-file HTML export

My least favorite thing about sharing converted Markdown used to be the images/ folder. You send the HTML, forget the folder, and the recipient gets a page full of broken image icons.

Here’s how the tool handles it: drop your .md files and your image files into the page together. The converter matches images referenced in the Markdown by filename, encodes each one as a base64 data URI, and embeds it directly into the generated HTML.

The output is one self-contained .html file. You can email it, drop it into a chat, or archive it, and the images display offline — no asset folder to zip alongside, nothing to fetch from a server later.

One trade-off worth knowing: base64 encoding makes embedded images take more bytes than the original files, so a screenshot-heavy document produces a noticeably chunkier HTML file. For the documents I share, that trade has been worth it every time.

Honest limitations

• The tool is client-side with one exception: the optional “import from URL” feature. Browsers can’t fetch arbitrary cross-origin pages, so that single path goes through a small proxy on my server. That is the only point where anything touches a server — if you never use URL import, nothing does. I’d rather say that plainly than have someone discover it in the network tab.
• It’s a converter, not an editor or a static site generator. It converts the documents you give it; it won’t build a site, navigation, or a theme around them.

Wrapping up

The link is up top if you want to poke at it. It’s free and there’s nothing to sign up for, so the cost of trying it is one browser tab.

If you convert Markdown to HTML (or back) regularly — what’s the thing that most often breaks for you? Footnotes, front matter, deeply nested lists, tables with inline code? I’d genuinely like to know what to test against next.