Will you write how this works or release the source?
When i receive such a request i use the object api http://develop.github.com/p/object.html to get a list of files in the repo, and then fetch *.textile and .png files which are new or have a new hash. I don't use their libraries, i just consume the rest interface directly.
I store the files in mongodb together with some metadata, and for processing i use a nice scala web framework called liftweb. For textile parsing i use http://www.eclipse.org/mylyn/downloads/ And for code formatting i use google-code-prettify
Anything else?