Apart from the SCORM Detective, here is a link to some sample content:
http://www.atutor.ca/view/16/4690/1.html
the "NATO" one, has the "problem" of HREFs being expressed as "./sco/whatever" which fools up the retrieval from the content repository (tried to modify the links in the manifest.xml, and load the course once again and it worked).
I guess this is a problem on our side and not of "poorly designed content". Therefore it should be addressed, but how ? My knowledge of the lors content repository is not that deep.
Does anyone has guidelines for such problems ? (which could be similar to the mixed case links as well as the non standard (accented) characters content).