Download websites that use jQuery CDN

About jQuery CDN

jQuery CDN is not a library, and treating the record as though it were leads straight to the wrong conclusion. What the entry describes is an edge in a graph: a page that fetches a script from a delivery network somebody else operates. The code that arrives is jQuery. The finding is about where the code comes from.

A site hosting its own copy and a site pulling the same file from a public network run identical code and carry entirely different risk. The second one has granted a third party the ability to serve executable code into its pages, has put a DNS lookup and a connection to a host it does not control on the rendering path, and has made its own availability partly contingent on somebody else's uptime. Stewardship of the library sits with OpenJS Foundation under the MIT license, and the delivery pattern dates from around 2006, old enough that a great many of these tags were written before anyone routinely thought about subresource integrity.

As of September 2026 we record the pattern on 2,409,123 sites and file it under CDN. United States accounts for 961,775 of them with .com carrying much of the volume, a spread that reflects where the tutorials and copy-paste snippets of the period circulated rather than any technical necessity, since the file is small and serving it locally has never been difficult.

The audience for a list like this is unusually specific. Supply chain security teams want every site that executes code from an origin its owner does not control, because the failure scenario for a delivery network is a single substituted script reaching an enormous number of pages at once. Content security policy work needs the same inventory, since each of these tags is an allowed external origin in somebody's policy and an argument the policy author has to defend. Performance consultants want it because a cross-origin fetch on the rendering path stopped paying for itself once browsers partitioned their caches, and the old justification that visitors would already hold the file from another site no longer applies.

None of those questions can be answered one domain at a time. A lookup tells you a single page references a delivery network, which a browser inspector would have shown faster and for free. The downloadable list answers the aggregate version: how many sites depend on a given third-party origin, how that dependency distributes across countries and domain suffixes, and which other technologies travel with it, which is what turns a curiosity into a risk register, a policy audit, or a market estimate.

The limits are worth stating plainly. Several networks serve the same library, so a single label answers whether a page pulls the library from a public network rather than which network it chose, unless a finer field appears in the export. Tags injected at runtime by another script can escape a crawler that does not execute everything a full browser would, which pushes the count down. And a page can reference an origin that no visitor ever reaches, because the tag sits on a template path nobody loads, so the record describes what a page declares about itself rather than what a real session necessarily fetches.