These 2 papers may be useful to you.

Horvath S (2013). "DNA methylation age of human tissues and cell types". Genome Biology 14 (10): R115. doi:10.1186/gb-2013-14-10-r115. PMC 4015143. PMID 24138928.

Fatemi M, Pao MM, Jeong S, Gal-Yam EN, Egger G, Weisenberger DJ, Jones PA (2005). "Footprinting of mammalian promoters: use of a CpG DNA methyltransferase revealing nucleosome positions at a single molecule level". Nucleic Acids Res 33 (20): e176. doi:10.1093/nar/gni180. PMC 1292996. PMID 16314307.

You can easily generate your own list. Download the CpG island track from UCSC:

https://genome.ucsc.edu/cgi-bin/hgTables

Choose the regulation group, CpG islands track.

You can then easily map these to gene TSSes, using whatever distance constraint you like best. Simplest web-based tool for site-to-TSS mapping is GREAT (but many other tools can do the job):

http://bejerano.stanford.edu/great/public/html/

Keep in mind that CpG island definition are fairly fluid - depending on the algorithm you use and the parameters you chose, you will see different island annotations. Some researchers do not count islands that have repetitive elements or at below a certain size. If these parameters are of interest to you, I would suggest doing de novo CpG island calling on your genome, using programs like newcpgseek (http://emboss.sourceforge.net/apps/cvs/emboss/apps/newcpgseek.html) or newcpgplot (http://emboss.sourceforge.net/apps/cvs/emboss/apps/cpgplot.html)

For the generic islands, you can download them from UCSC.http://genome.ucsc.edu/cgi-bin/hgTables

Use Genes and Gene Predictions - "known genes" for the gene annotations and Regulation -"CpG islands" for the islands, as mentioned above.

If you want direct overlap with a gene, you can use Galaxy (usegalaxy.org) to intersect these genomic intervals, using the 'intersect' tool and specifying how many bases of intersection you want.  You can also use tools to find the nearest non-overlapping interval.

We have created a CpG island annotation database using three CpG island detection algorithms, currently working on five organisms including human genome. According to a gene ID list as an input, it lets you observe all CpG islands found and to what proximity they are to the gene of interest. You can check our parameters more specifically through this link, if interested.

http://otulab.unl.edu/dbcgi