Clean Markdown Web Clipper

Markdown web clipper: save web pages as clean Markdown – no account, no sign-in. To clipboard, .md file, folder or Obsidian vault.

Markdown web clipper: save web pages as clean Markdown – no account, no sign-in. To clipboard, .md file, folder or Obsidian vault. Clean Web Clipper is a free Chrome extension that saves a web page as clean Markdown – to the clipboard, a .md file or an Obsidian vault. No account; the page text never leaves your browser. Works in Chrome on a computer; saving to a folder or an Obsidian vault asks you to pick the folder (again if access is reset). What does leave it: the address of a page that failed, the domain of a successful clip, the screen you opened, the clip action (no address) and a random install number. One switch in settings stops it all. It clips one page at a time, and a page with no article gets a plain refusal. GET A CLEAN NOTE • Menus, cookie banners and "read next" feeds are mostly cut. • Tables come out as Markdown tables; a small uneven two-column one, as key: value lines. • Code keeps the language tag the page declared, MDN and Sphinx too. • Footnotes become working [^1] references. • Title, author, date and source land in YAML frontmatter; with no date on the page the date line is left out, not set to today. SAVE WHERE YOU WORK • The clipboard • A downloaded .md file • A folder you picked on disk • Your Obsidian vault, with no plugin and without Obsidian running The icon click is yours to set: a preview window (the default), or saving straight away with no window. By default Alt+Shift+M copies the Markdown without opening the window. A To AI chat button copies the Markdown and opens claude.ai/new; it is in the menu, off on the footer by default. Handy for API docs and READMEs, papers with a real publication date, a reading list as notes, a regulation quoted word for word, web tables, and articles handed to an AI chat. SEE HOW IT WORKS The extension picks the article root by text density: the narrowest block that still holds almost all paragraph text outside links. Tables go through its own serialiser, so a cell with a list or code does not break the row. Folder and vault saving use the browser's File System Access API. If more than a quarter of the result is link labels, it answers "no article" instead of saving menu links; that page address is sent to us so the site gets fixed, and the Usefulness screen shows the last three pages it could not convert. It does not run on protected pages such as chrome:// and the Chrome Web Store. CHECK THE NUMBERS Tested on 512 top-site pages (August 2026): no crashes, no leftover HTML tags in articles. Method: https://lessroutinemorelife.com/clean-markdown-web-clipper/en/benchmark/ CONTROL WHAT IS SENT Page text, titles and notes stay in the browser. The five items above go only to our statistics server (via Cloudflare); never sold or shared for advertising. Addresses that look local (localhost, private IPs, .local, .lan and similar) are not sent. The install number is usually not linked to your visits to our website. The server records an approximate location (country, region, city) from your IP address. Turning the switch off erases the unsent queue and the install number. ASK BEFORE YOU INSTALL Can a clip be lost without me noticing? In a folder or vault an existing file is not overwritten: a number is added to the name, as in Article-2.md. Do I need an account? No account, no sign-in, no registration. Free, entirely. Website: https://lessroutinemorelife.com/clean-markdown-web-clipper/en/ Privacy policy: https://lessroutinemorelife.com/clean-markdown-web-clipper/en/privacy/ Support: https://lessroutinemorelife.com/clean-markdown-web-clipper/en/broken-site/ or cws@lessroutinemorelife.com