MediaIndexer works by storing metadata about the files in your repository in a SQLite database. Files are indexed by their md5 checksum, which allows them to be moved around within your repostory without having to update the database. The actual path to the file is not known by the database. The database only holds data such as a title, a description, user comments, number of accesses to the file, and it's mirroring URLs.

You may supply static content for each directory by placing specific files in their respective directories. This is to remain semi-compatible with Apache's Fancy Indexing feature. For example, the files HEADER.html, README.html, FOOTER.html are treated as a special files to MediaIndexer. If any of these files are found, they are printed as content for that page. This allows you to build an entire website with content around your file system.