the carbon with soric is not bad,
The problem with glueing is that any axial and angled force can breake the bond du to deformation in the bonded area.
Definetly MMA because it allows a bit of movement, Epoxy will just snap once the max force is exceeeded
As said in another post, if epoxy goes it goes with a bang, there are usally no warnings or signs of stress before that happans.

After all that all is subject tio testing, with a soric SF core, using rivets or bolts should not be a problem but need penny washers with a nylon between fibre and metal or backplate

Massimiliano, we are considering various stuff which may does not apply to you.

Just for fun I would glue one bracket onto a sheet, rest it horizontally and hang 4 x 25kg bags of sand or ballast on the bracket
I guess on a straight pulling force the bracket can hold at least 10 bags if your sheet is strong enough

Soric is a good choice, delamination is not very likely because of the honey-comb like resin filling bonding both skins.
I tested Soric with various fabrics and for stiffness and prefer EasyCell, but Easycell core deforms (stretches/compressed) under high load which will cause this delamination or break

The problem is shock, twisting and vibration which will exaggerate stress on the bonded surfaces, give MMA a go
Whats that part for ?

I just had a look at  Easy's MMA
it takes
stainless steel:   38.4N per mm2  -  30KG/mm2
GRP:                 Substrate Failure
 
You should be save to go ahead as long you don't hook an "elefant" on it

Araldite 2015 is not a bad choice here. Neither is an MMA. Using either of these adhesives, I would expect laminate failure before adhesive failure. Personally I would use the 2015.