It's confusing, but almost trivial. It's just a matter of definition of the Poisson bracket. I'm used to the definition
[tex]\{A,B\}=\frac{\partial A}{\partial q^j} \frac{\partial B}{\partial p_j} - \frac{\partial A}{\partial p_j} \frac{\partial B}{\partial q^j},[/tex]
where tacid summation over the index [itex]j[/itex] is understood (Einstein convention).
If you apply this definition to your example, you get
[tex]\{q^j,p_k \}=\delta_{k}^{j}.[/tex]
I could have defined the Poisson bracket as well with the opposite sign, and then I'd have gotten an additional minus sign on the right-hand side of the previous equation.
This is something one has to live with: Sometimes there are many equivalent definitions of the same mathematical structure, and when changing from one textbook to another or reading scientific papers you always have to check carefully, how the authors have chosen their convention.
This is most confusing in relativity, where it all starts with the sign of the "metric". In General Relativity you have also additional relative sign changes for the curvature, Ricci, and Einstein tensors. As I said, that's confusing, but one has to live with it unfortunately.