How to convert a URL to Markdown
- Paste any public URL into the box above.
- Press Convert. The tool fetches the page, extracts the readable article, and returns clean Markdown with the title, headings, links, and lists preserved.
- Copy the Markdown with one click and drop it into Obsidian, Notion, a prompt for ChatGPT or Claude, a blog draft, or any editor that understands Markdown.
What is a URL to Markdown converter?
A URL to Markdown converter takes the link to a webpage and returns the readable part of that page as clean Markdown. Markdown is a lightweight plain text format with a bit of syntax for headings, links, lists, and emphasis. It is the default format for AI prompts, most note-taking apps (Obsidian, Notion, Bear, Logseq, Roam), and a lot of documentation workflows. Converting HTML to Markdown throws out the nav, ads, sidebars, and JavaScript, and keeps just the article.
Why convert webpages to Markdown
- Prompting. LLMs handle Markdown better than HTML. Cleaner input gives better answers, uses fewer tokens, and runs faster.
- Notes. Paste a converted article into Obsidian, Notion, Bear, or Roam and keep the headings and links intact. Much better than pasting into a doc and getting a wall of text.
- Reading later. Markdown is easy to skim, archive, and diff over time. Saving articles as Markdown is the lowest-friction way to build a personal knowledge base.
- Writing. Quote or summarize an article without fighting with the source site's formatting.
How the conversion works
The tool runs a multi-stage extraction and HTML to Markdown pipeline:
- Fetch the URL with a safe server-side HTTP client.
- Detect the content type. Pages that are already Markdown or plain text skip the extraction step.
- Run the HTML through a readability extractor to find the main article body.
- Convert the extracted HTML into Markdown, preserving headings, links, lists, code blocks, blockquotes, and inline emphasis.
- Return the Markdown with the detected title and favicon.
What you get back
- Title. Pulled from the page's title tag or the extracted article heading.
- Body. The article content as Markdown. Headings, links, bullet and numbered lists, inline code, code fences, and blockquotes are preserved.
- Favicon. The site's favicon if one is advertised on the page.
- Source. Which extractor produced the result, so you can tell at a glance whether the site served Markdown directly or the content had to be extracted from HTML.
URL to Markdown for AI and LLM prompts
This is the most popular use case. Pasting raw HTML into an LLM wastes tokens and confuses the model with markup it does not need. Converting the URL to Markdown first strips the noise, keeps the structure the model actually cares about, and fits more of the real article in the context window. You end up with faster prompts, cheaper prompts, and better answers.
URL to Markdown vs HTML to Markdown vs webpage to Markdown
All three phrases describe the same job with slightly different inputs. URL to Markdown and webpage to Markdown take a link, fetch the page, and return Markdown. HTML to Markdown usually means you paste the raw HTML source and skip the fetch step. This tool handles the URL case. If you already have the HTML and just want the conversion, most Markdown libraries (turndown, html-to-md, pandoc) handle that path directly.
When a page will not convert cleanly
- JavaScript-only pages. If the article is rendered entirely by client-side JavaScript with no server HTML, the extractor has nothing to work with.
- Pages behind a login. The tool fetches pages anonymously, so private posts, gated articles, and intranet pages cannot be reached.
- Sites that block crawlers. Some publishers block anything that is not a mainstream browser. Those requests will fail the fetch step.
- Non-article pages. Home pages, search results, and product listing pages often do not have a single main article to extract.
Frequently asked questions
What does the URL to Markdown converter do?
You paste a URL. It fetches the page, pulls out the main article body, and returns clean Markdown with the title, headings, links, lists, and inline formatting intact. The rest of the page (navigation, sidebars, cookie banners, footers, ads) is stripped out so you get the content and nothing else.
How do I convert a webpage to Markdown?
Paste the URL into the box above and press Convert. The tool fetches the page, runs it through an extraction and HTML to Markdown pipeline, and hands back the Markdown. Copy it with one click and paste it into Obsidian, Notion, a prompt, or any other editor.
What is the difference between URL to Markdown, HTML to Markdown, and webpage to Markdown?
They usually mean the same thing. URL to Markdown and webpage to Markdown take a link and return Markdown. HTML to Markdown usually refers to pasting raw HTML source into a converter. This tool takes the URL, fetches the HTML for you, extracts the readable article, then converts that to Markdown.
Why would I want to turn a webpage into Markdown?
Three big reasons. First, feeding the page to an AI or LLM is much faster and cheaper with clean Markdown than with raw HTML. Second, Markdown pastes into Obsidian, Notion, Roam, Bear, and any notes app while keeping headings, links, and lists intact. Third, Markdown is portable, diffable, and future-proof in a way that a saved HTML snapshot is not.
Does it work for any page?
It works for most public articles, blog posts, docs pages, and news sites. It does not work well for pages that are entirely client-side rendered with no server HTML, pages behind a login, or pages that actively block scraping. When a page cannot be extracted cleanly, the tool tells you so rather than giving you noisy output.
Is there a URL to Markdown API?
Keep is the API. The tool on this page is a free web interface for the same extraction pipeline that powers Keep, a product that saves any URL as clean Markdown you can query from the dashboard, the extension, or your own code. If you are building something that needs URL to Markdown or HTML to Markdown at scale, Keep is the upstream version.
Does the Markdown include images?
Inline images from the article body are preserved as standard Markdown image syntax that points to the original source URL. Decorative images, tracking pixels, and images that belong to the site chrome are dropped along with the rest of the non-article content.
How long can the output be?
The free tool returns up to roughly 10,000 characters of Markdown per page. For most articles that is the full piece. Very long posts are truncated with a clear marker at the end. If you need uncapped conversion, longer articles, or batch URL to Markdown, save the URL in Keep and access the full clean Markdown from there.
Is anything stored?
The page you paste is fetched server-side so we can follow redirects and run the extraction pipeline, but the URL and resulting Markdown are not saved to a database or tied to an account. If you want a persistent, searchable library of clean Markdown pages, that is what Keep is for.
Save any webpage as Markdown with Keep
Keep is the bigger version of this tool. It saves every URL you bookmark as clean, searchable Markdown in a persistent library you can query, export, or hand to an AI agent. One URL now, the whole collection later.