[en] A two-step MCSCF procedure was developed based on the LDGUGA formalism. In one iteration the first step would be a CI calculation in which the orbitals were fixed. This is followed by a computation to determine how the orbitals should be changed to minimize the total energy. In this second step the orbitals vary but the expansion coefficients are frozen. The integrals are then transformed to the new orbital basis and we iterate until the changes in the orbitals and the expansion coefficients are suitably small. In computing the orbital changes the two-particle density matrix and the transformed integrals are needed. The ability to generate the two-particle density matrix rapidly is a key feature in our method