Plain Markdown

Changelog

1.7.2 #

Device limit + full-document AI for BYOK

  • Your license key now works on up to 5 devices. Activating it registers the browser with a random identifier (no browsing data), and devices you already use keep working. If you reach the limit, plainmarkdown.com/my-key gives you a fresh key that resets your devices
  • Bring-your-own-key AI now runs on the full document, limited only by your model's context window. The old 600,000-character trim only ever bounded costs on the server-powered Free and Pro tiers, which are unchanged
1.7.1 #

Sleeping-tab capture

  • Capture tabs now reads sleeping tabs too. Each is woken, captured, then put back to sleep
  • Fixed heading folding in the editor
1.7.0 #

Capture tabs

  • Capture tabs. Convert several open tabs into one Markdown document, all locally
  • Faster popup open (smaller startup payload)
  • Reading other tabs needs a one-time permission you grant in Settings
1.6.2 #

Image controls

  • Opening the extension never changes the page you're on
  • Keep or strip image tags in captured Markdown
  • Images fold into a clean pill in the editor
1.6.1 #

Multi-line AI prompts

  • Write multi-line AI prompts
  • Cleaner conversions, with more accurate author detection
  • BYOK: model search matches words in any order
1.6.0 #

Bring your own key

  • BYOK lifetime tier: run AI actions through your own OpenRouter key, with any model you choose
  • Set a default model and override it per action
  • Correct titles and content on single-page sites (Play Store, Amazon, Google search)
1.5.4 #

Notion integration fix

  • Connect Notion works again in published builds
1.5.2 #

Better extraction, cleaner Settings

  • Better extraction on Medium, LinkedIn, Bluesky, Threads, NYT, Wikipedia, ChatGPT, …
  • Lazy-loaded images, SVGs, and LaTeX captured correctly
  • Cleaner titles and authors (no template strings or placeholder leaks)
  • AI prompt limit raised to 10,000 characters
  • Settings → Plan refreshes AI usage on every visit
1.5.0 #

AI actions and destinations

  • AI actions: summarize, fact-check, …
  • Destinations: Obsidian, Readwise, Notion, …
  • Customize built-in actions or add your own
  • Refreshed Settings, empty states, and loading states
  • Pro tier: 100× more AI actions
1.4.0 #

Complex tables, smaller bundle, better extraction

  • Complex tables (colspan/rowspan) now convert to clean markdown instead of raw HTML
  • 40% smaller extension size
  • Better content extraction on more sites, including Substack, medical journals, and X/Twitter
  • Cleaner code blocks, footnotes, and link handling
1.3.1 #

Shadow DOM extraction fix

  • Fixed "No Content Found" on sites using shadow DOM (Reddit, Lit-based pages)
  • Upgraded content engine with improved footnote detection, code block rendering, and metadata accuracy
1.3.0 #

Mark folding and new extraction engine

  • Heading # marks fold away when the cursor is outside the heading line
  • Bold, italic, and strikethrough marks fold away when the cursor is outside the formatted range
  • Strikethrough (~~text~~) now parsed and styled
  • Floating panel renders at consistent size on all websites
  • Heading sizes adjusted to a more readable scale
  • Editor no longer auto-focuses on open, keeping all marks folded in reading mode
  • Ordered list numbers no longer dimmed (only bullet markers are muted)
  • Improved content extraction across a wider range of websites
1.2.1 #

Link folding

  • Link folding: inline link URLs are now hidden until you place the cursor on the link, dramatically reducing visual noise on link-heavy pages
1.2.0 #

Clipboard mode, skip images, and a new content engine

  • New "Copy to clipboard" open behavior: one click copies clean Markdown without opening a panel
  • New "Skip images" setting to strip image links from the output
  • Upgraded content engine: Shadow DOM support, improved content scoring, better footnote detection, CJK word count, and security fixes
1.1.0 #

Themes, settings, and the More menu

  • Added theme setting: dark, light, or system
  • Added open behavior setting: popup or floating panel
  • Added source info setting: include page title, URL, author, and date at the top of the output
  • Added More menu with Re-run, Download, and Settings
  • Improved code block and inline code styling for better readability
  • Fixed link contrast in light mode
  • Fixed tables missing from some converted pages
  • Fixed list items showing a fold arrow in the editor
  • Fixed author detection: no longer falls back to site name when no author is found
1.0.1 #

Cleaner output, full page conversion

  • Cleaner Markdown output across more websites
  • Improved reliability for complex pages
  • Full page conversion is now the standard for a simpler, more predictable experience