Loop to Markdown

Tampermonkey userscripts (or bookmarklets) that convert Microsoft Loop pages to Markdown or MediaWiki with one click.

Features

• Tables, code blocks, headings, lists (ordered, unordered, checkboxes)
• @mentions, hyperlinks, inline code
• Bold text detection
• Code language auto-detection (Python, JavaScript, SQL, YAML, JSON, Mermaid, etc.)
• Loop page title detection and export as the document title
• Table of contents export as a nested list
• Quip link capture
• Clipboard fallbacks for environments where GM_setClipboard is unavailable

Available Scripts

• loop-to-markdown.user.js: exports Loop pages as Markdown
• loop-to-mediawiki.user.js: exports Loop pages as MediaWiki wikitext

Installation

1. Install Tampermonkey browser extension
2. Install the exporter you want:
   • Markdown userscript
   • MediaWiki userscript

Bookmarklet (no extension required)

Prefer a bookmarks-bar button over a browser extension? The same exporters are also available as bookmarklets in dist/.

1. Download or clone this repo and open dist/install.html locally in your browser (GitHub shows the file as source rather than rendering it)
2. Drag the button you want onto your bookmarks bar:
   • 📋 Copy Loop as Markdown
   • 📋 Copy Loop as MediaWiki
3. Open a Microsoft Loop page and click the bookmark to copy the page

The bookmarklets are generated from the userscripts by build-bookmarklets.py, so the two stay in sync — edit a *.user.js, then run python3 build-bookmarklets.py to regenerate dist/.

Usage

1. Open any Microsoft Loop page
2. Click the exporter button in the bottom-left corner (userscript) or the bookmarks-bar bookmarklet:
   • 📋 Copy as Markdown
   • 📋 Copy as MediaWiki
3. Paste the copied output into your target editor or wiki

Notes

• Markdown exports add the detected Loop page title as # Title when available.
• Loop reserves heading level 1 for the page title, so section headings keep their Loop levels (## and below).
• Collapsed and virtualized code blocks are expanded and scrolled through during export so their full text is captured; the page scrolls briefly while this happens and is restored afterwards.
• Both scripts try GM_setClipboard, then the browser Clipboard API, then a document.execCommand('copy') fallback.

Contributors

• Talha Oz - Original author
• Yuta TJ (yutatj@) - Table cell fallback extraction
• Shrinivas Acharya - Bold detection, code language detection, ordered lists
• Andrea Scian - Clipboard fallbacks, Loop title detection, heading offset, MediaWiki exporter

Contributing

Found a bug or have an improvement? PRs are welcome! This script was built with contributions from multiple people - yours could be next.

License

GPL-3.0