[[*gʰeud-]]
Fragment of a discussion from User talk:Rua
The gullet entry gives *gʷel- (which we also don't have an entry for) as PIE root.
I agree with CodeCat here about sources. From what you've said, the Adams book doesn't appear to say that "gut" is derived from *gudom; the sources should explicitly make the link. What does Pokorny say, for example? PIE *g -> Germanic *g is unexpected. gut vs. gudom could be a coincidence (cf. have vs. Latin habēre). Note also that etymonline agrees with the derivation from *gheud (although this is not a primary source).