Webpage to Markdown Converter (URL to Markdown)
Paste a public URL and get the page as clean Markdown: article text, headings, links and tables kept; scripts, navigation and boilerplate dropped. Nothing is stored.
You want the content of a page, not the page. A help-center article to rewrite, a policy page to send to legal, a competitor's pricing table to put in a deck, a product page to hand to whoever is writing the macros. Save-as gives you two megabytes of HTML. Select-all and copy gives you the navigation, the footer, the cookie banner and a newsletter box, with the article somewhere in the middle.
Paste the URL of a public page. It is fetched once, script, style and other non-content blocks are stripped, and the rest is converted to Markdown on Cloudflare's document-to-Markdown service. Navigation and repeated boilerplate are dropped; headings, paragraphs, lists, links and tables stay. No model is involved, and nothing is kept beyond a short cache of the result. This is the format an agent reads — the same conversion powers WireDesk's knowledge ingestion.
This ran once, on one input. The product runs on your whole site, every conversation.
Convert your whole help center into an agent — start freeWhat each input means
- 01
Paste the URL
The full address, including https://. It has to be a page anyone can open without signing in: help centers, docs, policy pages, blog posts, product pages. A page behind a login, a paywall or an IP allowlist returns the login page or an error.
- 02
Convert to Markdown
Press the button. The page is requested once from Cloudflare's network, so the site sees one visit from Cloudflare rather than from you. It's converted in memory and discarded. Scripts on the page are removed, not run.
- 03
Copy the result
Copy to the clipboard. Skim it once: boilerplate removal is a heuristic, so a bit of footer occasionally survives and a sidebar with real content is occasionally cut. If the result is nearly empty, the page is rendered by JavaScript — see the FAQ.
Questions about this tool
No. The page is fetched as an anonymous visitor, so anything behind a login, a paywall, SSO or an IP allowlist comes back as the login page or an error. For those, open the page yourself, save it as HTML or copy the source, and use the HTML to Markdown converter, which never needs to reach the site.