How to get a page's HTML
Browsers don't let one website read another website's pages, so paste the page source in here instead. Any of these work:
- View source: open the article, press
Ctrl+U(Mac:Cmd+Option+U), select all, copy, and paste it below. - Save the page: press
Ctrl+S, choose "Webpage, HTML only," and open the saved file here. - Developer tools: press
F12, open the Console, runcopy(document.documentElement.outerHTML), and paste. This captures content that scripts added after the page loaded.
You can also paste plain text, such as text copied from a PDF or an email, and it will be reflowed into clean paragraphs.
Paste a page's HTML or some text above, and the clean version will appear here.
Clean up any web page for distraction-free reading
News sites, recipe blogs, and documentation pages bury the actual text under menus, cookie banners, ads, share buttons, newsletter pop-ups, and "you might also like" rails. Readability Cleaner finds the article inside all of that and shows only the parts worth reading: the title, author, date, body text, headings, images, lists, quotes, code blocks, and tables. You can then adjust the text to read comfortably, print it without the clutter, or save it as HTML, Markdown, or plain text.
The cleaning happens entirely in your browser. The HTML you paste isn't uploaded, and none of the page's own scripts run, so trackers and pop-ups in the source can't do anything.
How to use it
- Copy the page's source (see "How to get a page's HTML" in the tool for three quick ways) and paste it in, or open a saved
.htmlfile. - Optionally paste the page's address so relative links and images point to the right place.
- Choose what to keep. Turning off images or links gives you the leanest version for printing or archiving.
- Pick a theme, font, size, and width that suit you. The tool remembers these for next time.
- Copy the result, print it, or download it.
What gets removed and what stays
The cleaner works in three passes. First it strips anything that can't be part of an article: scripts, styles, forms, embedded frames, navigation bars, sidebars, footers, and hidden elements. Then it drops blocks whose class names or roles mark them as ads, social widgets, comment sections, related-post lists, or cookie notices. Finally it scores what's left. Paragraphs earn points for length and punctuation, those points flow up to the containers holding them, and link-heavy blocks lose points. The highest-scoring container, plus closely related neighbors, becomes the article.
| Removed | Kept |
|---|---|
| Menus, headers, footers, and sidebars | Title, site name, author, and publish date |
| Ads, sponsored blocks, and promo boxes | Paragraphs, headings, and lists |
| Share buttons and "follow us" bars | Images with captions (optional) |
| Related stories and recommendation rails | Links inside the text (optional) |
| Comment sections and newsletter sign-ups | Block quotes and pull quotes |
| Cookie banners, pop-ups, and overlays | Code blocks, keeping their spacing |
| Scripts, trackers, styles, and forms | Data tables (optional) |
Every tag in the result is rebuilt from a short list of safe elements, and every attribute except link addresses, image sources, and table spans is dropped. That's why the output looks consistent no matter how the original page was styled.
Cleaning plain text
Text copied from a PDF, an email, or a terminal often arrives with a line break at the end of every line. Paste it in and the cleaner rejoins the lines into proper paragraphs, repairs words that were hyphenated across a line break, turns lines that start with dashes, bullets, or numbers into real lists, and drops stray boilerplate lines like "Advertisement" or "Share this article."
Download formats
- HTML saves a single self-contained page with simple reading styles. It opens in any browser and can be printed or converted to PDF later.
- Markdown is ideal for note apps like Obsidian and Notion, static site generators, and anything you'll edit later. Headings, links, emphasis, lists, quotes, code, and tables all carry over.
- Plain text is the smallest and most portable. Lists keep their markers, and tables become rows separated by vertical bars.
Copy places both a formatted and a plain version on the clipboard, so pasting into Word or Google Docs keeps headings and bold text while pasting into a plain editor gives clean text. Print sends only the article to the printer, with no site header, sidebar, or ads.
Frequently asked questions
Why can't I just enter a URL?
For security, browsers block a web page from reading content on a different site unless that site allows it. Reading pages by URL would require a server to fetch them for you, which means your reading history passes through that server. Pasting the source keeps everything on your own computer.
Do images load from the original site?
Yes. When "Keep images" is on, the pictures are shown from wherever the page stored them, so the image host sees a request just as it would if you visited the page. The tool asks the browser not to send the page address along with it. Turn images off if you want nothing to load at all.
The result is missing part of the article. Why?
Some sites load the article text with JavaScript after the page opens, so it isn't in the plain page source. Use the developer tools method in the tips box, which copies the page as it currently appears, including script-added content. Articles split across several pages also need each page cleaned separately.
Will it work on paywalled articles?
It only works with HTML you already have. If a site sends the full article to your browser and hides it with an overlay, the cleaner will usually show it. If the site never sends the text, there's nothing to clean.
Is it safe to paste HTML from unknown sites?
Yes. The source is parsed as an inert document that can't run scripts or load files, and the reading view is rebuilt from scratch using only safe tags. Links open in a new tab with the referrer removed.
Can it clean a Word document or PDF?
Not directly, but you can copy the text out of either one and paste it here. The plain text mode fixes the broken line wrapping that PDFs usually produce.