I guess LC-MS would be safer since under GC conditions at elevated temperature, alkaloid decomposition would be unavoidable. An LC-MS equipped with PDA (DAD) woulb be better option. However, I would also consider the fact that many alkaloids of same class or family happen to have same MS-spec (same MW) even though they are completely different compounds. In that case only MS would be misleading. Start with TLC (look for positive spots for alkaloids: Dragendorff's or other Cerium based staining agent) followed by Chromatographic separation to confirm the structure by NMR and MS.
"the fact that many alkaloids of same class or family happen to have same MS-spec (same MW) even though they are completely different compounds."
That would not be a problem at all and that is why MS provides you with fragmentation data so you can easily detect different compounds with the same Mw. Plus the GC and LC dimension of retention will help to resolve them apart.