There is not currently a way to extract that value. It is something I could add relatively easy (I believe) for the next release of Spike.
In the meantime, if you are retrieving the I/Q data, you can make this measurement with the equation
10*log10(mean(I^2 + Q^2));
In spoken word, calculate the linear power of each I/Q sample, I^2 + Q^2, calculate the mean of the linear powers, then take the 10*log10 of that mean.
This would give you the average power in dBm over the sample interval of your choice.
Given this, would you still be interested in a SCPI command to retrieve this value?
I look forward to your response.