Clean up Markdown source in one click: add spaces between CJK and Latin text, insert the missing blank line before lists, align table columns, renumber ordered lists and unify heading and bullet styles — code blocks are left byte for byte. Pasting from Word or a web page converts to Markdown automatically; copy the result or download it as .md.
Upload a Markdown file, preview it locally in your browser and export it to Word or PDF. Nothing is uploaded.
Wondering what opens a .md file? Drop a .md / .markdown file here and read it right in your browser, nothing to install. View-only: no editing, no exporting, no uploads.
A Markdown file is just plain text: # marks a heading, - marks a list item, and ** marks bold. It opens fine in Notepad — you just see the raw text with its symbols rather than a laid-out page. Because it is plain text, the files are tiny, Git can diff them properly, and they will still open in ten years even if some particular application stops being maintained. That is exactly why developer documentation, open-source READMEs and more and more note-taking apps use it. The price is that when you double-click one, Windows usually has no idea what to open it with.
This is the fork in the road. If someone sent you a README or exported notes and you just want to read them, use the viewer: drop the file in and it renders immediately, with no edit mode, so the question "might I accidentally break someone else's file" never comes up. If you need to hand it on — to a colleague who only has Office, as an attachment, or to a printer — you want Word or PDF. The third case is a file that is simply written badly: a list jammed against the paragraph above it with no blank line, table pipes that do not line up, CJK and Latin text crammed together. That calls for the formatter, which rewrites the source itself and hands you back a .md file.
Markdown has a few whitespace-sensitive spots where getting it wrong raises no error and simply renders as something else. The classic one is a list pressed right up against the paragraph above it with no blank line — under CommonMark the whole list is treated as a continuation of that paragraph, so the - shows up literally. A heading whose # is missing the space after it is the same kind of problem: fine in some editors, no longer a heading once it is pushed to GitHub. Faults like these are usually invisible in your own editor and only surface when the file is opened somewhere else, which is why formatting fixes more than appearances.
Heading levels, lists, bold and italic, links, tables, block quotes and images all survive intact. What you do lose is mainly syntax highlighting in code blocks (in Word they usually become plain monospaced text) and some Markdown extensions, such as task-list checkboxes and footnotes. Math formulas do come across, but in Word each one is an image and can no longer be edited; if the document is heavy on formulas, a PDF export normally stays closer to the original than a Word one.
Parsing Markdown and generating Word / PDF can both be done entirely inside the browser; not one step of it has to happen on a server. The benefit is not only privacy: with no upload there is no file-size limit, no queue and no "you have used up today's free quota", and the page keeps working with the network disconnected. Work documents, drafts that are not published yet, internal material — none of it has to be handed to a server you know nothing about first.