adamsmasher,
cool proof of concept. i'm going to be the nay sayer. i set out to build my first hack in august and there are few things that i have done that have been as frustrating. the #1 reason: every guide (every one) that i read was outdated. screen shots of multibeast and unibeast don't match the current versions. names of things have changed. version control and authoritative sourcing of required files is simply out of control. "did you try scott's version of the blue.kext? it replaces gary's version of the blue.kext." there are thousands and thousands of posts on what to do, but a general void of the whys. my hardware now runs Win7 because i wanted to build a mac to use. i have no interest in hackintoshing for a hobby.
all of this to say that if this is going to be a spare-time project, there is simply no way that the guidebeast will be able to 1) stay current 2) have sufficient, useful information. life will be better for you if you high five all around for the POC and move on to the next cool idea. however, if this is something that is the target of your passion and you are committed to keeping it current, then consider an editorial board that is comprised of authorities for each section. define the scope of the content before proceeding. define persona who will use the guidebeast. give each persona a profile. what do they know? what do they need to know? first build? 4th build? looking for basic, step-by-step, or in-depth instructions. without this, the likelihood of a generally useful guide soon is low.
this is a BIG undertaking. really big.
hope this helps.