Your font produces shapes that correspond to Unicode code points. If the characters you want to display are not (yet) in Unicode, you can use Private Use Area code points, but then the data will only make sense if that font is used.
Keyman produces Unicode code points according to what keyboard is being used. It can produce PUA code points if the keyboard is written to generate them.
If the code points produced by the keyboard match the code points supported by the font, then the character shapes should appear. (That’s a simplified summary of the process, but the point is that the connection between the keyboard and the font is the code points that you choose.)