Best Fit a "Skewed" Normal Distribution

[Lente]

Hello,

First time poster, long time reader.

I'm stumped, and am looking for some guidance.

I am attempting to use a few data points to generate a "best fit" skewed normal distribution from which i can extrapolate other data.

Inputs:
1 target value at some point on the curve.
Value that would be considered to occur 95% of the time in a random sample (2 std dev to the left of mean on a normal curve)
Value that would be considered to occur 5% of the time in a random sample (2 std dev to the right of mean on a normal curve)
% 'probability' between target value and the left value
% 'probability' between target value and the right value.

If I've been unclear, let me know; I'd be happy to clarify.

Thanks