One of the significant objective of geophysical data recording and processing is the enhancement of the signal-to-noise ratio, and in this perspective, the design of an optimum digital filter is pivotal. The well-known Wiener-Hopf filter have been successfully applied to attain this objective, and the output energy filter is such tool, by which one can attempt to enhance the signal-to-noise ratio by retrieving the signal at the output, producing longer burst of energy in the time interval where the signal occurs. In the present work, we report the development of a digital filter, namely, gravity energy filter for the improvement of gravity signal-to-noise ratio that immersed in coloured noise. For the design of such filter, it is not obligatory to have explicit knowledge of the gravity signal shape; nevertheless, its performance level is not compromised. We demonstrated its applicability on synthetic data, generated by considering two spherical bodies at the subsurface, and hence immersed in coloured noise which has led the gravity signal to be obscured. We found that our tool has successfully applied on the synthetic gravity data, and the gravity signal buried in coloured noise beyond visual recognition is easily detected after filtering.