Copy as Markdown 作者： Bruno Volpato
Context-aware "Copy as Markdown" button — the fastest way to share web content with LLMs like ChatGPT, Claude, and Gemini.
9 个用户9 个用户
扩展元数据
关于此扩展
Copy as Markdown adds a smart floating button to every website. One click copies the page content as clean, structured Markdown — perfect for pasting into ChatGPT, Claude, Gemini, or any LLM.
🧠 WHY MARKDOWN?
When you copy-paste from a website, you lose all structure — headings become blobs, tables vanish, links disappear. Screenshots eat your token budget. Manually reformatting takes forever.
Copy as Markdown preserves everything: headers → #, tables → pipes, code → fenced blocks, links → text. The result is 90%+ smaller than raw HTML while keeping full semantic structure.
🎯 24 FIRST-CLASS EXTRACTORS
Purpose-built extractors for the most popular sites:
• Wikipedia — article body, infoboxes, tables
• GitHub — issues, PRs, repos, README, code
• Stack Overflow — questions, answers, vote counts
• YouTube — title, description, chapters, transcript
• Reddit — posts, threaded comments with depth
• Hacker News — title, score, comment threads
• arXiv — papers with title, authors, abstract
• Medium — articles, author, claps, reading time
• Substack — newsletter posts, paywall detection
• Dev.to — articles, tags, reactions, comments
• MDN Web Docs — documentation, code examples, browser compat
• ChatGPT — shared conversations with role labels
• NPM — package info, README, dependencies
• PyPI — Python packages, version, description
• Google Search — query, snippets, knowledge panel
• Google Docs — full document export
• Bing Search — results, knowledge sidebar
• Amazon — product title, price, rating, reviews
• LinkedIn — profiles, posts, articles
• X (Twitter) — posts, replies, engagement stats
• Polymarket — market odds, resolution rules
• WhatsApp Web — chat messages with timestamps
• News sites (20+) — CNN, BBC, NYT, Reuters, and more
🌐 WORKS ON EVERY WEBSITE
Not on the list above? No problem. The smart fallback extractor works on any website — it detects the main content area (article, main, or selection) and converts it to Markdown automatically.
✨ KEY FEATURES
• One-click copy — floating button appears on every page
• Selection-aware — highlight text first to copy just that section
• YAML frontmatter — adds source, title, URL, and date metadata
• Dismiss per page — click the ✕ to hide the button for the current session
• Zero data collection — everything runs locally, nothing leaves your browser
• Lightweight — single JS file, no external dependencies
📋 PERFECT FOR
• Sharing web content with AI assistants (ChatGPT, Claude, Gemini)
• Research and note-taking in Markdown editors
• Developers copying documentation and code references
• Content curation and knowledge management
Open source: https://github.com/bvolpato/copy-as-markdown
🧠 WHY MARKDOWN?
When you copy-paste from a website, you lose all structure — headings become blobs, tables vanish, links disappear. Screenshots eat your token budget. Manually reformatting takes forever.
Copy as Markdown preserves everything: headers → #, tables → pipes, code → fenced blocks, links → text. The result is 90%+ smaller than raw HTML while keeping full semantic structure.
🎯 24 FIRST-CLASS EXTRACTORS
Purpose-built extractors for the most popular sites:
• Wikipedia — article body, infoboxes, tables
• GitHub — issues, PRs, repos, README, code
• Stack Overflow — questions, answers, vote counts
• YouTube — title, description, chapters, transcript
• Reddit — posts, threaded comments with depth
• Hacker News — title, score, comment threads
• arXiv — papers with title, authors, abstract
• Medium — articles, author, claps, reading time
• Substack — newsletter posts, paywall detection
• Dev.to — articles, tags, reactions, comments
• MDN Web Docs — documentation, code examples, browser compat
• ChatGPT — shared conversations with role labels
• NPM — package info, README, dependencies
• PyPI — Python packages, version, description
• Google Search — query, snippets, knowledge panel
• Google Docs — full document export
• Bing Search — results, knowledge sidebar
• Amazon — product title, price, rating, reviews
• LinkedIn — profiles, posts, articles
• X (Twitter) — posts, replies, engagement stats
• Polymarket — market odds, resolution rules
• WhatsApp Web — chat messages with timestamps
• News sites (20+) — CNN, BBC, NYT, Reuters, and more
🌐 WORKS ON EVERY WEBSITE
Not on the list above? No problem. The smart fallback extractor works on any website — it detects the main content area (article, main, or selection) and converts it to Markdown automatically.
✨ KEY FEATURES
• One-click copy — floating button appears on every page
• Selection-aware — highlight text first to copy just that section
• YAML frontmatter — adds source, title, URL, and date metadata
• Dismiss per page — click the ✕ to hide the button for the current session
• Zero data collection — everything runs locally, nothing leaves your browser
• Lightweight — single JS file, no external dependencies
📋 PERFECT FOR
• Sharing web content with AI assistants (ChatGPT, Claude, Gemini)
• Research and note-taking in Markdown editors
• Developers copying documentation and code references
• Content curation and knowledge management
Open source: https://github.com/bvolpato/copy-as-markdown
评分 5（1 位用户）
权限与数据
更多信息