why registry? I think the reason is that reading the registry is a known easy thing to program in & most consoles have default script that you can just plug in key to read & report back.
Here is an approach that might come in handy. Small, silent utility that is software deployment friendly.
The reason we write to the registry is that we already are capturing registry data using another package. We just added these new values to the data being collected so that we can then do reporting based on the resultant values.
Handy if you're trying to do a "Delayed Provision" - and need to know what systems are Remote Configuration capable and have not been provisioned\configured.
Some environments - like Altiris - use a client based agent and "discovery" to determine some of the core AMT components and states.
Curious though - why place the information in the registry?