Markdown editor

GitHub-flavoured Markdown: headings, emphasis, lists, links, code blocks, quotes, tables, task lists, strikethrough and links written as plain addresses. Single line breaks are joined, as GitHub shows a file. Footnotes, maths and diagrams aren’t rendered.

The preview doesn’t load anything from other sites: loading an image would tell that site your address. Each one shows as its description and address instead. Images embedded in the Markdown as data: addresses are shown.

The HTML is cleaned before it’s shown: scripts, event attributes, frames, forms and styles are removed, and links other than http, https, mailto and # are turned into plain text. The copied HTML and the downloaded page are the cleaned version, and the page carries a policy that stops it loading anything from the network.

Only if you turn on “Keep draft in this browser”. Then it’s kept in this browser’s storage (up to 1 MB) and comes back when you open the page again; turning it off deletes it. Otherwise nothing is kept, and the browser asks before you leave with changes you haven’t downloaded. It asks too if the draft is on but couldn’t be kept.

Over 500,000 characters the preview pauses, because showing it would hold up the page for seconds; “Update preview” shows it when you want. Copying and downloading the Markdown work as usual; the HTML ones wait for an up-to-date preview.

No. It’s rendered in your browser, and a file you open is read there too. It isn’t sent to our servers or put into the page address.