Download websites that use jQuery

About jQuery

jQuery has been declared obsolete more times than most libraries have been released, and as of September 2026 we still detect it on 39,243,137 sites. The distance between those two statements is the whole reason this page is worth reading.

The obituaries were not wrong about the technical argument. The gaps jQuery was written to paper over, inconsistent browser APIs and hostile selector engines, were closed years ago by the browsers themselves. Nothing built today needs it. That is a statement about new work, and new work is a small fraction of the web at any given moment.

What keeps the number high is everything already shipped. We currently see 1,161 distinct versions in production, with 3.5.1 the most common. Old releases persist because thousands of themes, plugins, page builders and one-off agency builds pinned a version at the moment they were written and were never revisited. Each of those is a small decision that nobody has revisited, and in aggregate they form a population that changes far more slowly than the discourse around it.

This is why presence alone is close to worthless as a signal here. Nearly every site of a certain age carries it, so a filter on presence returns most of the web. The useful cut is always by version, because that is where the actionable difference lives. A site running a release from before a known advisory is a materially different prospect from one running a current build, whether the context is a security assessment, a maintenance pitch or a due diligence exercise.

Version data also doubles as an age estimate. A library pinned years ago is a reasonable proxy for when a site was last meaningfully rebuilt, and that inference tends to be more reliable than anything the site says about itself. Combined with the CMS and the hosting underneath, it produces a picture of how actively a property is maintained, which is the question behind a large share of commercial research.

Distribution skews toward United States, where we detect it on 14,332,690 sites, and toward the .com zone, both of which follow the age profile of those markets rather than any preference for the library itself.

The webatla export carries the version string for every domain we detect it on, alongside the other technologies on the same site, as one file rebuilt daily. That structure exists specifically so the version filter is cheap: you are not paying per lookup to discover that most of the answers are uninteresting, you are filtering a full population locally until the interesting subset falls out.

The scale also makes it a useful control variable rather than a target. Because it is present across such a wide cross-section, filtering on it barely narrows anything, which means it can be used to test whether some other signal in your analysis is real or merely tracking site age. If a correlation you have found disappears once you hold this constant, you were measuring age.

There is one more reading of the version data that tends to get missed. Clusters of sites pinned to exactly the same old release usually share an origin: a theme, a plugin, a page builder or an agency template that bundled that version and shipped it everywhere. Those clusters are discoverable only by grouping the full population by version and then looking at what else the members have in common, which is a two-step operation over a local file rather than a query anyone offers as a service.

It is distributed under the MIT license by the jQuery Foundation, and has been public since 2006. Nobody has ever had a list of who uses it, because there has never been anything to register with.