What an agent 'ought' to do, according to Gewirth under the principle of generic consistency, does or would 'protect the rights [of others] to freedom and well-being'. But I don't think 'ought' can in Gewirth's view be defined in terms of such providing such protection.
My understanding of Gewirth (admittedly a complex thinker, so I write with less than complete confidence) is that his 'ought' is a moral requirement that we respect the freedom and well-being of others on pain of inconsistency given our own status as voluntary and purposive agents.
Plainly an argument is needed to get to this requirement from the mere fact of our being such agents. Gewirth tries to provide this. He takes voluntariness and purposiveness as two generic (or as he sometimes says, 'categorical') features of human action.
... in explanation Gewirth quickly adds, "By 'voluntary' I mean that
the agent occurrently or dispositionally controls his behavior by his unforced choice, knowing the various proximate circumstances of his action.
By 'purposive' I mean that the agent intends to do what he does, envisaging
some purpose or goal which may consist either in the performance of the
action itself or in some outcome of that performance: in either case insofar
as it is the purpose of his action the agent regards it as some sort of good."
Given these categorial or, if you will, basically factual features of all human
action, Gewirth feels that any agent has but to engage in such action, and
he is thereby committed to an acceptance of the moral point of view; or, to
put it a little differently, what Gewirth insists upon is that the 'is' factor of
such action is implicative of an 'ought' factor as well.
But how so? The implication, Gewirth thinks, may be made clear in
three steps. First, by the very fact of his acting thus voluntarily and purposively, any agent is committed, at least implicitly, to a judgment to the
effect that his own freedom and well-being are things of an undeniable
worth and value to him. There then follows the second step, which consists
in the agent's being committed to the further judgment that, his own
freedom and well-being being matters of such fundamental moment to
him, he must regard them not just as matters of mere appeal and attraction
but as matters of right as well. That is to say, he makes the "corresponding
right-claim" that he is "justified in performing his actions and in having
the freedom and basic well-being which generally figures in all his actions." No sooner, though, is such a right-claim made than the third step
follows, which amounts to an invocation and application of the principle of
universalizability: if a man's acting voluntarily and purposefully is a matter
of right for him, then it is no less a right for any and every man; and
therewith a moral order of reciprocal rights and duties comes to pervade the
entire scene and to implicate the agent, as it were willy-nilly, in the moral
point of view. (Henry B. Veatch, 'Paying Heed to Gewirth's Principle of Categorial Consistency', Ethics, Vol. 86, No. 4 (Jul., 1976), pp. 278-287: 280-1.)
What, though, of the soundness of such a derivation of a moral 'ought'
from the 'is' of such categorial features as are necessarily associated with
any human action? Seemingly, the first and the third steps of the derivation
are unassailable. But what of the second step? Does the mere fact of an
agent's placing an unmistakable value upon his own freedom and well-
being necessarily involve his making a right-claim to freedom and well-
being as things that should be his as a matter of right? Is it not easy enough
to imagine someone who does indeed place not just value but perhaps
supreme value upon his own freedom and well-being, but who never for a
moment claims such things to be due him or to be owing him, it being
rather his determination to make them his merely through his own efforts
and, if need be, by a most ruthless competition and conquest? Particularly if
such a Thrasymachan or Hobbesian or Nietzschean individual were to
equate 'freedom' with freedom to do as he pleases and 'well-being' with a
limitless wealth and power over others, then why need there be any claim
of rights of any kind, not to mention duties? (Veatch: 284.)