Not sure if that is of any help, but I remember that I had a small modification of of bond/react to accept lammps variables as input to the reaction probability. But this was a per reaction and not a per atom value if I remember correctly. I can look that up if necessary.
Regards
Wolfgang