Web page to Markdown converter
Paste a link and imread turns the web page to Markdown: the article with its headings, lists, tables, links and code, and none of the share buttons, captions or stray links a direct copy drags along. Copy the Markdown in one click. Free, no sign-up.
What does the Markdown keep from the web page?
The converter fetches the page, finds the article and drops the site around it. What stays is the structure a reader would care about:
- Headings keep their levels, so the outline of the article survives.
- Lists stay lists, ordered or not.
- Tables with real data come out as Markdown pipe tables. Tables used only for page layout are read as ordinary paragraphs.
- Links keep their destinations, so the sources an author cites are still one click away.
- Code stays in code blocks instead of collapsing into running text.
Menus, sidebars, cookie banners, share buttons and related-post grids are left out. That is the difference between this and selecting the whole page and pressing copy: a direct copy takes the interface with it, and the Markdown here is the article alone.
How do I convert a web page to Markdown?
Paste the address of the page into the box above and choose Fetch. After a few seconds the article appears as Markdown. Press Copy to put it on your clipboard, then paste it into your notes app, a document or a prompt. Choose Open in reader if you would rather read the page first in imread's reader; there, the arrow beside Copy offers Copy as Markdown as well, so the same result is one step away.
New to the reader? The guide to using imread walks through the other controls: type size, listening and translation.
Markdown or plain text: which should I copy?
Both come from the same page, and the right one depends on where the words are going.
- Markdown suits notes apps, documentation and prompts where structure matters. Headings, lists, links and code blocks arrive intact.
- Plain text suits a chat message, an email or a word-processor document. It is clean paragraphs, and tables come out as tab-separated rows, so they paste into a spreadsheet as columns.
If you cannot decide, the comparison of Markdown and plain text checks a real article export for what each format keeps, including source links and code. For plain text alone, use the link to text tool.
Can I paste the Markdown into ChatGPT or my notes?
Yes, that is the usual reason to convert a page. A language model reads Markdown well: headings tell it where sections begin, and a table stays a table instead of a stream of numbers. Pasting the article as Markdown also leaves out the navigation text that would otherwise use up the space of your prompt and distract from the question you are asking.
Notes apps that understand Markdown render it straight away: headings become headings and links stay clickable. For a quick summary, a research folder or a reading list, paste the Markdown, add your own notes under it and keep the original address, which the result carries with it.
Does it work on WeChat, GBK and SEC pages?
These are the pages that usually fight back, and imread is built to read them:
- WeChat Official Account articles can be converted to Markdown from their public link. To read one in English, see how to translate a WeChat article.
- Chinese and Japanese pages in older encodings, such as GBK and Shift_JIS, are decoded correctly. Other tools turn them into garbled characters.
- SEC EDGAR filings come back with their tables intact, as Markdown pipe tables.
How do I convert a web page to Markdown from a script?
The same Markdown is one request away, and no key is needed:
curl "https://imread.org/api/extract?url=https%3A%2F%2Fpaulgraham.com%2Fgreatwork.html&format=markdown"The response is the article as Markdown, with the address URL-encoded in the url parameter. AI agents can call the same conversion as a tool through imread's MCP server. The API docs cover long pages that arrive in parts, and the errors you may meet.
What can the converter not read?
The fetch asks for the page the way an anonymous visitor would, and it reads the HTML the server sends. That sets honest limits:
- JavaScript-only pages. The extractor does not run scripts, so a page that builds its content in the browser may come back as an empty shell.
- Logins. The fetch cannot sign in, so pages that need an account give back only what a logged-out visitor sees.
- Paywalls. A paid article returns its public excerpt or a notice, not the full text.
- Blocked sites. If a site refuses machine access, the fetch fails with a message that says so.
Free use is limited to a small number of pages a minute from one device. A very large page is cut short rather than refused.
Questions about converting a web page to Markdown
Is the web page to Markdown converter free?
Yes. There is no sign-up and no key for the page above or for the API request shown earlier, within the free rate limit.
Does it keep images?
The conversion is aimed at the article's words: its headings, lists, tables, links and code. Do not count on the Markdown to carry the article's pictures; open the original page when you need them.
Why is my Markdown short or empty?
Most often the page builds its content with JavaScript, sits behind a login or a paywall, or blocks automated requests. Open the link in a private window: if you see less than the full article there, the converter will see the same.
Can I get Markdown without opening the website?
Yes. Use the curl request in the script section above, or let an agent call the MCP server. Both return the same Markdown that the Copy button gives you.