It depends on the gamma attenuation coefficient of the medium. Refer shielding textbooks.

My simple guess is, when prompt gamma travel through reactor shielding, only high energy group of gamma will be detected after the shielding, never knew a single traveling gamma can actually loose energy like a neutron slowing down in moderator.

No, @Mohamad,

all gammas can be detected outside the shielding with reduced intensities, there is no 100% absorption of photons. Indeed you should have a look at the photon interaction processes to get an idea about the different interaction modes of photons. The attenuation coefficient depends on density, Z and A of the absorber and from the energy of the photons.

A typical rule is, low energy photons are prefered for attenuation, so spectra are "hardened", the medium energy increases with depth in the medium. But new photons will be found (annihilation quants 511  keV) and fluorescent radiation from the electron shells of the passed atoms and molecules.

Especially there is no stopping power for photons or neutrons like for charged particles where photons or neutrons  loose their energy by retardation in small portions. The moderating process of neutrons is caused by elastic collisions with other nuclei, the only interaction partner of the unloaded neutrons.

Addendum to Hanno Krieger's answer:

Study the subject "Build-up-factor" in the shielding textbooks.

The prompt fission gamma spectrum does not depend on  the travelling of  the emitted gammas, it can be treated in the calculations as a well-defined function of the gamma-energies (and a lot of other parameters, e. g. the fission channel). Due to the absorption of the low-energy fraction, the spectrum ("far" from the place of the fission) would be harder than the source spectrum. 

On the other side, the gammas scattered out from the high-energy part will be present in the low-energy part. The importance of this phenomenon is small in short distances from the source but increasing in larger path lengths. Of course, the distance must be measured in the energy-dependent mfp-s.

The problem  described in M. H. Rabir's original question is similar to the one solved by the very old MERCURE-3 shielding code (3D multigroup source, straight line attenuation to the detecting point with build-up-factor correction)

Actully，only when we know the core gamma rays flux distribution, then it is possible to evaluate the γ irradiation impact on core components.

Yes, all the guys are right. It heavily depends on the characteristic of the material the gamma ray would travel through.