If you are injecting the typical 1-2 uL sample volume at the split ratios you describe then there should not be damage to the common polysiloxane stationary phases. CHCl3 has a boiling point of about 61 C, so you be somewhat limited to separating compounds with higher boiling points.
If you're using a standard polysiloxane column you shouldn't have a problem - water's the one you really want to avoid, so if you're doing biomass extractions, make sure your sample is dried.
Ideally, in my opinion, such polar extracts should not be analyzed in GC/MS if not derivatized. Because such solvents can extract many undesirable compounds from the target matrix. e.g. if you extracting aerial parts like leaves. there may a possibility to have chlorophyll in your sample if not treated with active charcoal. These may get accumulating at column injector end. Similarly, several other non volatiles may also stuck at the injector and column. These can lead to loss in number of theoretical plates responsible for better separation. e.g. a 30m length x 0.2 mm id PDMS column possessed 4500 plates per meter. Hence, more the plate available better will be the separation. Absence of this may lead to retention gap.
I agree with Andrea's comments if you inject small microlitre with good split ratio. Definitely, water damages capillary column as suggested by Tom in his comments.
If you are targeting halogenated compounds in your samples, the isotopic ratio pattern of halogenated compounds in mass spectrometer can help you in understanding the number of halogens present. Moreover, for quality separation in GC, you must have an ECD detector.
If you wants to carry on your analysis in GC/MS only, then I will advice you to installed a 2-5m guard column before the actual PDMS or PEG columns to avoid any kind of damage.
You may have HPLC, a best option for such analysis.