How Skip the Paywall Works: Complete Guide and Troubleshooting

Skip the Paywall is deliberately simple: one box, one button. But a lot happens between pressing "Read Article" and seeing the text, and when something goes wrong it helps to know what the tool is doing. This is the complete guide: how it works, what each message means and what to try when an article will not load.

What happens when you paste a link

• The link is checked. It has to be a full web address starting with http or https. If you paste a bare domain without a path, the reader will fetch the homepage, which is rarely what you want.
• The page is fetched from our server, not your browser. The request identifies itself the way a search engine crawler does, with a Google referrer, because that is how publishers expect indexing traffic to arrive. No cookies, no login, no subscriber credentials.
• If the site refuses, the reader looks for a saved copy in the Internet Archive and fetches that instead.
• The article is extracted with Mozilla's Readability, the same open-source library behind Firefox's reader view. It finds the main article on the page and discards navigation, ads and sidebars.
• The result is cleaned: scripts, embedded frames, forms and tracking pixels are removed, image addresses are made absolute, and the text is shown to you.

Nothing is stored. Each read is generated on demand and discarded, and the reader pages are blocked from search engines so that they never compete with the original article.

Which sites work

Any site with a metered or freemium paywall that is enforced in the browser: most magazines, most regional newspapers and platforms like Medium. The full list we have tested is in metered versus hard paywalls, and the home page shows the current set of supported sites. Sites with server-side paywalls (Financial Times, Wall Street Journal, Bloomberg, The Economist) and sites with bot protection (The New York Times, The Telegraph, The Atlantic) do not work, and our post on sites that cannot be bypassed explains what to do instead.

The error messages and what they mean

"Unable to fetch the article. The site may be blocking automated access, or the URL may be incorrect."

The publisher's server refused the request, timed out, or returned an error, and no archived copy was found. Causes, in order of likelihood:

• The site blocks automated requests. This is the normal result for The New York Times, The Atlantic, Forbes, Bloomberg, The Telegraph and similar. There is no fix; see the alternatives post above.
• The site is slow. The reader gives up after eight seconds so that it can show you an error rather than hang. Very heavy pages occasionally exceed that. Try again; second attempts often succeed because the publisher's cache is warm.
• A typo in the link. Check that the URL opens in a normal browser tab.
• The Internet Archive is down. The archive fallback only helps when the archive is reachable, and it has outages.

"Could not extract article content. The page may not contain a readable article."

The page loaded, but the extractor could not find an article in it. This usually means the link points at something that is not an article: a homepage, a section front, a video page, a gallery, a live blog or a search result. Find the specific article link and try that.

The article loads but is only a paragraph or two

The site sent a teaser rather than the full text. This is what a server-side paywall looks like from the outside, and it is the normal result on The Wall Street Journal, The Economist, Barron's and The Washington Post. No reader can get more than the site sends.

The article loads but with menu text at the top or bottom

Some sites, notably the Postmedia papers in Canada, wrap their articles in so much navigation that the extractor keeps a line of it. The article itself is complete; scroll past the stray labels.

"Too many requests. Please wait a minute and try again."

The reader limits each visitor to ten articles a minute to keep the service free for everyone. Wait sixty seconds.

Things that are not articles

The reader extracts text and inline images. It does not handle:

• Video and audio players, which are removed
• Embedded tweets, gists and social posts, which are removed as frames
• Interactive graphics, data tables loaded by script, and maps
• PDFs and print replicas
• Anything behind a personal login, such as a subscriber-only app or an e-edition

Tips for the best results

• Paste the direct article link, not a shortened link, a Google AMP link or a link from a newsletter tracker. Those often redirect through a page the extractor cannot read.
• On a phone, set up the one-tap bookmark so you do not have to copy and paste.
• If a site that used to work starts failing, it has probably changed its paywall or added bot protection. We re-test the supported list regularly and update the home page.

Where to read the fine print

Our post on the legal and ethical questions explains how we think about all this and where we draw the line. The short version: we fetch public pages, we store nothing, and we would rather you subscribed to anything you read every day.