Gamma jets (GJ) at MeV energies may be produced in galactic halos by I
nverse Compton Scattering (ICS) of collinear ultrarelativistic electro
n beams (at GeV energies) onto thermal photons. These electron jets ar
e emitted by neutron star jets (NSJ) or black holes (BH) while the tar
get thermal photons are generated by a nearby binary stellar-like comp
anion (or by a relic accreting disk). The GJ precesses in a conical sh
ape as a lighthouse (with a characteristic Keplerian binary period) fo
rced by the NSJ magnetic lines interaction with the stellar companion
ones. The sudden blazing of the GJ, once the observer is into the beam
cone direction, is detected as a GRB. Our predicted GRB spectra fit t
he GRB observed ones; GRB luminosity, variability and counts are well
explained if the GJ sources are located in an ''extended'' and an ''ev
aporating'' galactic halo as the one recently required in the fast run
ning away NS model. Observed jets such as SS433, the Great Annihilator
, superluminal sources such as the last GRS1758-258, GRS1915+105 are t
he ideal candidates for NSJ as well as GJ. The SGRs are the lower ener
gy and wider beamed version of this GJ model so they are linked to cla
ssical GRBs. The last mysterious presence of the twin rings near SN198
7A (as well as the ring traces near NGC6543) are the spectacular signa
ture of such (double sided) conical jet precession projected onto the
red giant relic spherical shell. The tiny but very puzzling offaxis na
ture of the SN1987A twin rings is due to the high eccentricity of the
binary NSJ. The GRBs from the nearby Andromeda halo might be detected
as soon as the instrumental sensitivity is increased.