We study the phenomenological viability of a constrained NMSSM with parameters subject to unified boundary conditions from F-theory GUTs. We find that very simple assumptions about modulus dominance SUSY breaking in F-theory unification lead to a predictive set of boundary conditions, consistent with all phenomenological constraints. The second lightest scalar Higgs H_2 can get a mass m_{H_2} ~ 125 GeV and has properties similar to the SM Higgs. On the other hand the lightest scalar H_1, with a dominant singlet component, would have barely escaped detection at LEP and could be observable at LHC as a peak in H_1 -> gamma gamma at around 100 GeV. The LSP is mostly singlino and is consistent with WMAP constraints due to coannihilation with the lightest stau, whose mass is in the range 100-250 GeV. Such light staus may lead to very characteristic signatures at LHC and be directly searched at linear colliders. In these models tan(beta) is large, of order 50, still the branching ratio for B_s -> mu+ mu- is consistent with the LHCb bounds and in many cases is also even smaller than the SM prediction. Gluinos and squarks have masses in the 2 - 3 TeV region and may be accessible at the LHC at 14 TeV. No large enhancement of the H_2 -> gamma gamma rate over that of the SM Higgs is expected.