[plt-scheme] Using DrScheme as editor for an external Scheme system

[I tried finding info about this specifically but was unsuccessful.  I'd
be thankful to anyone for pointing me to relevant info.]

What I'd like to do is make a new DrScheme langauge for an external
Scheme system running as a separate OS process that allows me to use
DrScheme's (very nice :-) editing of the "definitions window" and
"interactions window" and have those windows' contents sent to the
external process and have the external process's output shown (probably
after I've parsed and filtered it a little) in DrScheme.

I already know how to use PLT Scheme's functionality to run the process
and communicate with it and all that, but how would I make a DrScheme
language for this integrated with DrScheme's editing of buffers and
REPL?

I'm wondering how practical it is to attempt this?  

PLT Tools: DrScheme Extension Manual, 2.1.2 says: "With some additional
work, any language that can be compiled to MzScheme's language is
supported by the tools interface, not just those that use standard
configurations and module."

Should I keep digging into this document and review the DrScheme
languages others have made, or should I take a different approach?  Do I
need to be aware of any caveats for what I'm attempting?  

Should I look at what's coming in version 400+ which might change how my
idea would be done?

Thank you very much for any help on how I could do this.

-- 
: Derick
----------------------------------------------------------------