What does the URL to Markdown API do?

It fetches a web page, removes navigation, ads, scripts and other boilerplate, extracts the main content, and returns clean Markdown — ready to feed an LLM, RAG pipeline, or agent. Free, no API key, open CORS.

Yes. GET a URL, get Markdown, no signup or key. For high-volume or commercial use, contact us for a Pro tier.

Any web page → clean Markdown

Feed a page to an LLM without the nav bars, cookie banners and ad slots. This API fetches a URL, strips the boilerplate, and returns clean Markdown — the main content only. No key, open CORS.

Free. GET a URL, get Markdown.

Try it

curl "https://urltomd.rootnest.ai/md?url=https://example.com"

Run it in your browser →

API

GET/md?url=<page> — returns text/markdown

GET/md?url=<page>&format=json — returns { url, title, markdown, chars }

JSON example

fetch("https://urltomd.rootnest.ai/md?url=" + encodeURIComponent(page) + "&format=json")
  .then(r => r.json())
  .then(d => console.log(d.markdown));

Open CORS — call it from the browser, a script, an agent, or your build step. Only http/https HTML pages; 12s fetch timeout.

Why this exists

Every RAG pipeline and agent needs the same thing: a web page as clean text, not a soup of nav, scripts and cookie prompts. Doing it well means handling readability extraction, relative-link resolution, and markdown formatting — the fiddly parts you don't want to rebuild. This API does that one job, so your agent gets clean context and you don't maintain an HTML-cleaning stack.

Building an AI product with this? For higher volume, batch endpoints or a commercial license, email hello@rootnest.ai.

Notes

Best-effort readability extraction (prefers <article>/<main>). Not affiliated with any site you convert; respect the source's terms and robots. Trademarks belong to their owners.