We are trying to understand the undocumented functions and features of this tool.

Our understanding: The repository finds files in a given site (URL address).
Our findings: sometimes it finds files, sometimes not, some types are found, others not.

Some examples:
http://eledia.de/
Moodle logos are found.
Header pic http://eledia.de/sites/all/themes/eledia/images/start.jpg is not found
Linked PDF http://www.dialoge.net/dok/eledia/hosting.pdf is not found

Generally it seams that graphic files can be found, but not other file types. If you add the URl to an other file type directy into the field its been found (tested for pdf pptx and mp4 files

An other example:
http://www.friseur-michels.de no result after search
The page contains a jpg file <a href="home.html"><img src="bild/logo/Marcel Michels Logo Banner.jpg" alt="Friseur Bonn Marcel Michels" width="482" height="61" class="Mitte"></a>
May be the spaces in file name is a problem

The videos at <source src="http://www.friseur-michels.de/video/Collection2011.mp4" type="video/mp4">
<source src="http://www.friseur-michels.de/video/Collection2011.ogg" type="video/ogg"> are not found.

Can you please explain whats a feature and what a bug.

Generally it doesn't make any sense to implement feature without any type of information or documentation.