I think LC-MS would be more suitable for these compounds, but if you want to analyze them by GC-MS you should first hydrolyze (by acidic, alkaline or enzymatic hydrolysis) your material in order to obtain the corresponding saponigens. After that, go for silylation to enhance volatility of your compounds: add Pyridine-BSTFA-TMSCl (ratio 1:4:1) to a sample of your material in a test tube, add cap, put the silylated sample in an oven at 70 degrees Celsius for 30-45 min (shake the test tube after 10 minutes) for sufficient silylation, transfer into a vial and then run your analysis. The oven temperature could be 150 to 270 C at a rate of 10 C/min and then holding it at 270 C for 10 min. This is just a rough estimation, though. Several things can affect this (e.g., column length, column health, flow rate of carrier gas and so on). Good luck.

I forgot to reply to the question in the title. In my opinion, GC-MS alone isn't enough for characterization. NMR on pure, isolated&individual saponins accompanied with LC/GC-MS would be the best characterization "tool".

For Hydrolysis can i use hydrochloric acid..? And for methylation BF3 Methanol..?

Yes, HCl would be fine. BF3 in MeOH I haven't used since we mainly use trimethylsilyl groups as derivatization, but I know it is widely used for methylation of fatty acids (even if there is a risk of cis-trans isomerization of double bonds in conjugated fatty acids!). But if you have carboxylic functions in your molecules, why not.

FAMEs analaysis as per AOAC 996.06 Standard  you  can go  by  BF3 Derivataion  is  the  best  one method