A web-to-Markdown converter that keeps the article, not the page
Most web-to-Markdown tools just dump the HTML and call it a day. The result is full of cookie banners, sidebars, sticky footers, ad slots, and navigation menus, all of which distract your AI. Ours runs Mozilla’s Readability extraction first, then converts only the meaningful content area, and puts back any data table the extraction left behind.
You get a clean Markdown document with the page’s own headings, real lists, links made absolute so they still work outside the site, and image references. Run it on one URL, on a list of URLs, or crawl a site from its start page: a practical way to build AI knowledge bases out of public docs, blogs, and reference sites.
What gets stripped
- Navigation menus, headers, footers, and side rails
- Cookie banners, modals, and newsletter pop-ups
- Ads, scripts, styles, forms and embedded frames
- Comment sections and related-posts widgets
What gets preserved
- Article headings, paragraphs, lists, and quotes
- Code blocks with language hints
- Tables, images, and inline links
- Page metadata as an optional YAML frontmatter block: URL, title, description, author, site name, image and language
Pages are read as the server sends them, without running JavaScript. A page that builds its content in the browser comes back empty or nearly empty.