(Ori, I hope you are not offended by our confusion. I assume that English is not your native language and I assure you that your English is far better than my command of whatever language
is your native language!)
I'm not at all sure what "stocks" are here and I THINK that "vector rotor" is the curl. I'm pretty sure this person is using "S" to indicate integral and the problem is to integrate the vector function (xz^2-y)dx+(3x-yz^2)dy+(zx^2-zy^2)dz
around the intersection of the surfaces given by x^2+y^2=1 and z=xy.
Aha! Stoke's theorem! x
2+ y
2= 1 is the cylinder along the z-axis. We can write that in parametric equations as x= cos(θ), y= sin(θ), z= z (with θ and z as parameters) and the intersection of that with z= xy is x= cos(θ), y= sin(θ), z= sin(θ)cos(θ).
Integrating the vector function around that curve would be tricky but doable. To use Stoke's theorem you need to find the curl of the given vector function.
You don't really need to find an expression for a surface inside that boundary. The nice thing about Stoke's theorem is that it applies to ANY surface having that boundary.
Take as your surface the surface z= xy itself. Here is how I would do it:
f(x,y,z)= xy- z= 0 has z= xy as a "level surface": div f= yi+ xj- zk is perpendicular to that surface and we can "normalize" to the projection onto the xy-plane by dividing by the k component: Integrate the dot product
((xz^2-y)i+(3x-yz^2)j+(zx^2-zy^2)k).(-y/z, -x/z,1)dx dy.
You will, of course, need to use z= xy to reduce the integral to x,y only. The integration is over the unit disk so you may want to convert to polar coordinates.