If the authors compressed the JAR files into a single bundle, storing the data into a BrowserFS-created IndexedDB file system removes decompression/extraction overhead on subsequent visits.
If they didn't.... then it might be sufficient, but I am not sure how well the browser cache handles large files! How long until it evicts them? Is there an upper bound on the size of cached items? etc.