The possibility of testing the presence of new resonances in the Large Hadron Collider (LHC) data has attracted a lot of attention after the preliminary results indicating possible excesses in the Run 1 data by both ATLAS and CMS Collaborations in the diboson channels, including also the more recent diphoton excess at 750 GeV. In the following I discuss the possibility that a weak singlet pseudo-scalar particle, η WZ , decays into two weak bosons after being produced in gluon fusion at the LHC. The couplings to the gauge bosons arise in this case from a Wess-Zumino-Witten anomaly term which is calculable explicitly (not a free parameter) in models where the pseudo-scalar arises as a composite state, as the coefficients of the anomalous couplings can be related to the fermion components of the underlying dynamics. I provide simple examples of these composite realisations. When such hints will be tested in more detail in the Run 2 analyses of the LHC data it will be possible either to pinpoint in more detail the structure of the composite electroweak sector or to give bounds on it.