.MD this page is a browser extension that extracts the main content of an article or webpage, removes clutter, and converts it into well-formatted Markdown. Whether you're saving articles, preparing content for large language models, or building knowledge bases, this extension helps you turn messy web pages into structured, readable text in seconds
Hello Product Hunters,
".MD this page" is a browser extension that extracts the main content from any article or webpage, removes clutter like ads and navigation, and converts everything into clean, well-structured Markdown with a single click. It’s especially useful for working with modern LLMs, which perform far better when given organized Markdown instead of noisy HTML resulting in clearer structure, better reasoning, and more efficient use of tokens. The extension uses smart extraction (powered by Mozilla Readability), offers a dedicated preview interface, and lets you customize the output by toggling images, links, metadata, and even generating a page structure map. You can then easily copy the result, download it as a .md file, or use it directly as a prompt for AI workflows—turning messy web pages into clean, LLM-ready documents in seconds.
The extension is open source:
https://github.com/Ademking/MD-T...
Report
What does it use under the covers to convert the html to markdown? Some pre-existing library or rolls its own? Is there LLM assistance at all?
This is super practical — I've been manually copying and cleaning up web content for my LLM workflows, and it's always a pain dealing with ads, navbars, and random HTML artifacts. Having a one-click 'convert to clean Markdown' is exactly the kind of utility tool that saves cumulative hours. The open-source angle is great too. Does it handle pages with heavy JavaScript rendering well, like SPAs or dynamically loaded content?
For JS-heavy pages: it works as long as the content is already visible on the page when you click. So most sites are fine.. For heavy SPAs or lazy-loaded stuff, you might need to scroll or wait a bit first
Report
Can this be done programmatically? Is there a library that can convert webpages to Markdown so it can be used by agents?
RepoList
What does it use under the covers to convert the html to markdown? Some pre-existing library or rolls its own? Is there LLM assistance at all?
RepoList
@gunnarniels The pipeline is basically: rendered DOM → Readability extraction → Turndown → Markdown
No LLM involved at all.. everything runs locally in the browser
@adem_kouki Love it
This is super practical — I've been manually copying and cleaning up web content for my LLM workflows, and it's always a pain dealing with ads, navbars, and random HTML artifacts. Having a one-click 'convert to clean Markdown' is exactly the kind of utility tool that saves cumulative hours. The open-source angle is great too. Does it handle pages with heavy JavaScript rendering well, like SPAs or dynamically loaded content?
RepoList
@new_user___093202624e1848eb9b9896f Really appreciate that.. that’s exactly the pain I had too
For JS-heavy pages: it works as long as the content is already visible on the page when you click. So most sites are fine.. For heavy SPAs or lazy-loaded stuff, you might need to scroll or wait a bit first
Can this be done programmatically? Is there a library that can convert webpages to Markdown so it can be used by agents?
RepoList
@koushikg Yes, you can use the Turndown library