The plasma environment of comet 67P provides a unique laboratory to study plasma phenomena in the interplanetary medium. There, waves are generated which help the plasma relax back to stability through wave–particle interactions, transferring energy from the wave to the particles and vice versa. In this study, we focus on mirror-mode-like structures (low-frequency, transverse, compressional and quasi-linearly polarised waves). They are present virtually everywhere in the solar system as long as there is a large temperature anisotropy and a high plasma beta. Previous studies have reported the existence of mirror modes at 67P, but no further systematic investigation has so far been done. This study aims to characterise the occurrence of mirror modes in this environment and identify possible generation mechanisms through well-studied previous methods. Specifically, we make use of the magnetic-field-only method, implementing a B–n anti-correlation and a new peak/dip identification method. We investigate the magnetic field measured by Rosetta from November 2014 to February 2016 and find 565 mirror mode signatures. Mirror modes were mostly found as single events, with only one mirror-mode-like train in our dataset. Also, the occurrence rate was compared with respect to the gas production rates, cometocentric distance and magnetic field strength, leading to a non-conclusive relation between these quantities. The lack of mirror mode wave trains may mean that mirror modes somehow diffuse and/or are overshadowed by the large-scale turbulence in the inner coma. The detected mirror modes are likely highly evolved as they were probably generated upstream of the observation point and have traversed a highly complex and turbulent plasma to reach their detection point. The plasma environment of comets behaves differently compared to planets and other objects in the solar system. Thus, knowing how mirror modes behave at comets could lead us to a more unified model for mirror modes in space plasmas.