If you include ribosomal and/or mitochondrial genes in your transcript list (such as RefSeq genes), those will be among the genes with normalized read counts.
Different people will have different definitions for "housekeeping" genes. Genes like GAPDH should be present in pretty much any commonly used transcript list.
Off the top of my head, the only thing that could be used for QC purposes that wouldn't normally be included are spike-ins (like ERCC). However, those have to be added as part of the wet lab protocol - they won't automatically be there. In fact, if the web lab protocol includes ERCC spike-ins, then I would assume the core / company providing the sequencing should automatically include those sort of statistics.
If you are more broadly asking how to quantify RNA expression levels, these are probably the most common strategies (all of which should have prepared reference files for human and mouse):
1) TopHat (or STAR, etc.) + cufflinks
2) TopHat (or STAR, etc.) + RSEM
3) Bowtie (or BWA, Novoalign, etc.) + eXpress
Here are the links for some of the programs listed above: