I have several circles under the same circle group/node
the taged should be like the circled couple, but for ALL
I have no idea about how to approach this problem, sorry I am still learning 
to build a circle you need the Plane on which the circle lies and a radius: donāt know how convenient is to create a circle on a plane and then rotate the circle itself, it sounds better to me to rotate the planes on which the circles get directly created
the thing of having the circles as output of a given component doesnāt really tell much about the data structure they are organized in 
please get used to internalising data and post a GH file with the geometries of interest, even if it might look silly for such a small definition 
here I flattened the center point list (so the data structure is lost):
you could Unflatten the list of circles at the very end if the data structure of the center points is a thing you want to maintain through the definition
thanks, but like this, it doesnt actually check if adjacent planes are oriented in the same orientation, or Iām wrong?
May I know how does your ālist itemā node have 3 outputs vs mine who has only one?
you can zoom-in more with the mouse wheel
when the component becomes big enough, some small +/- will appear on output top and bottom
clicking those +/- will allow you to increase the number of outputs
that does not work on all components, but there are a few that let you customize the number of inputs/outputs (Merge, Entwine, List Item just to name a few)
Here is a link to the Grasshopper Primer:
you are right, it just relies on the random triplets seeds to generate different-enough solutions
the concept of āintroducing checksā could be simplified in generating a rotation value within a Range of possible values, all of which would make you equally happy
that would -maybe- simplify the process?
itās like picking a random fruit from a basket of fruits knowing you donāt like half of those, or just putting in the basket a smaller selection of fruits all of which you like 
but you really need to explain more of what you are trying to achieve
As always, thanks
itās actually an xy problem
So as the previous thread, if I extract the wireframe of an object, then āpipe itā through sweep1, in some locations I get those flat or almost flat surfaces,
which I thought I may be able to get rid by having the circles, or rather the planes as you told me, in uneven rotations, which by the way, they also seem to persist even after a bool difference.
ps: I did quite enjoy the modelab guide, I wish to have learned its existance sooner, thanks
have you tried the Pipe component?
it creates a pipe (circular section) of chosen radius around a curve, and takes care of the section-circles to be always perpendicular to the curve
