Download websites that use Flask

About Flask

Flask began as an April Fools joke. The first release wrapped two existing libraries, Werkzeug and Jinja, into a single file and presented the result as a parody of the microframework fashion of the day, and enough people asked to use it seriously that the parody turned into a maintained project. The original gag, that a whole framework could fit in one file with a decorator for routing, is still roughly how the thing works.

Flask is a small Python framework maintained under the Pallets Projects umbrella and released under BSD. Flask first appeared in 2010 with a deliberately narrow core, routing and templating and not much else, leaving database access, authentication and form handling to separate extensions. The design made it a default starting point for APIs, internal tools, research demos, dashboards and the service layer behind mobile applications.

Detection of a server-side framework rests on indirect evidence. Headers from the server process in front of the application, the shape of session cookies, default error pages, static path conventions and occasional framework-specific behaviour are the signals available from outside. A site placed behind a CDN, a reverse proxy or a managed platform can remove all of them. Our count is a lower bound and should be read that way, and the bias runs in a predictable direction: the deployments we can see lean toward smaller and less hardened setups, while large production estates are the ones best at staying quiet.

Versions behave the same way. We record 45 versions in circulation, with detections clustering around 3.1.8. Python applications are usually deployed with pinned dependencies and then left alone until something forces an upgrade, which is why a live service can sit several releases behind for years. For dependency risk work the version column is the part of the row that matters, because the framework itself changes almost nothing about how a page looks.

As of October 2026 we detect Flask on 3,402 sites. A single lookup answers whether one domain uses it. A downloadable extract answers the questions that only exist in aggregate: which hosting providers concentrate Python applications, how usage splits by country and domain ending, what else those sites run alongside the framework, and which parts of a market are built by developers rather than bought as a package. Python tooling vendors, hosting and platform companies, security teams and recruiters looking for engineering-led organisations are the readers who need the whole list rather than a sample of it.