Of course, you are right that 'there is such a thing as a "NAMD topology for AMP."' I meant to say a CHARMM topology file for AMP.
Unfortunately, the CHARMM topology files I have (and I have looked through each of them extensively) do not explicitly define a residue type called "AMP", so I'm trying to decide what the best course of action is for "inventing" such a definition to adequately describe adenosine monophosphate. So that's my problem at hand.
As for the previous poster, that's a good idea to check out what some forums specifically oriented toward NAMD have to offer in this regard. It's sure worth a try!