There are more than 1.6 billion websites today, and almost every one of them uses JavaScript libraries. Knowing that, it’s very important to show problems that occur as a result of not paying enough attention to security, such as using outdated versions of JavaScript libraries, insecure libraries, and so on. This paper proposes an algorithm for JavaScript library version detection. The algorithm detects version of JavaScript libraries based on differences between neighboring library versions. It’s designed in such a way that it can be run periodically and automatically on a server. The paper also presents results and efficiency of the algorithm on a smaller set of data collected from the Croatian Web space. The success of the algorithm in detecting the correct version is about 50%, and the range of probable versions is an additional 25%. From these results, i.e. the detected versions, we found that the JavaScript libraries used on the websites of the Croatian web space are not regularly updated. Limitations and also possible potential improvements to the algorithm are listed at the end of the paper.