CDO

You are right the parameter angle is missing in the documentation. The default is 0.

Okay. I still need clarification on the rotation.

1. Is the globe rotated such that the north pole traces a great circle from its original position to xnpole,ynpole?
2. And then rotated counter-clockwise around the new north pole be angle degrees?

Thanks

We don't have data on a rotated grid which are using the rotation angle and I'm not familiar with those grids. As far as I know xnpole and ynpole is simple the geographic position of the new pole. The angle needs to have the unit degree but I don't know the direction of rotation.

Imagine the new pole being at 0E,0N. That fact says nothing about where the prime meridian is. The prime meridian could be on any great circle between 0E,0N and 180E,0N.

Fishing through the CF Metadata Mailing List I found:
There is one model in the CMIP5 database (inmcm4) that has a rotated-pole
grid (for the ocean) which uses a non-zero north_pole_grid_longitude. This
is the first time I've encountered the use of that parameter, so I'm glad
that we included it in the CF definition! It was me, I think, who gave it that
name, and I'm sorry if it's unclear. It seemed fairly clear to me; the idea
is that it's the grid_longitude of the true North Pole.
I guess that the angle is the same as the CF Metadata north_pole_grid_longitude

Yes CDO stores the angle to the netCDF attribute north_pole_grid_longitude. But I'm not sure whether this is correct or not.