Ieke Moerdijk has written a small Springer Lecture Notes tome addressing this question:"Classifying Spaces and Classifying Topoi" SLNM 1616.
Roughly the answer is: A G$G$-bundle is a map whose fibers have a G$G$-action, i.e. are G$G$-sets (if they are discrete), i.e. they are functors from G$G$ seen as a category to Sets$\mathsf{Sets}$. Likewise a C$\mathcal C$-bundle for a category C$\mathcal C$ is a map whose fibers are functors from C$\mathcal C$ to sets$\mathsf{Sets}$, or, if you want, a disjoint union of sets (one for each object of C$\mathcal C$) and an action by the morphisms of C -$\mathcal C$ — a morphism A-->B$A \to B$ in C$\mathcal C$ takes elements of the set corresponding to A$A$ to elements of the set corresponding to B$B$.
There is a completely analogous version for topological categories also.