If we use the "origin-based" definition of gamma rays (i.e. photons emitted from nuclear decay, regardless of wavelength), which decay produces the lowest-energy, longest-wavelength emission? This comment Are the Properties of the EM Spectrum Fluid? alludes to low-energy gammas in what would usually be considered the ultraviolet band; are there or could there be any in the visible?
There is no lowest energy. There is no lower bound on the difference between nuclear energy levels, and in fact for hot nuclei the density of levels grows exponentially. Between any two nuclear energy levels, there is in principle the possibility of an electromagnetic decay. In reality there are various difficulties involved in detecting very low-energy gamma rays, and they tend not to compete very well against higher-energy gammas.
In addition to Ben Crowell's answer:
Thorium 229 has a nuclear isomer Thorium 229m that has an energy 8.28±0.17 eV above the ground state. When it decays it does so via an UV wavelength photon (which is a honorary gamma by its origin). As far as I know this is the lowest energy emission known to regularly occur.