Tameness conjecture for $\mathcal{O}(n)$-geometries

Let an $\mathcal{O}(n)$-geometry be a Cartan geometry modelled on the corresponding model, and call it tame according to the source's definition. Tameness conjecture. An $\mathcal{O}(n)$-geometry is tame if and only if it is isomorphic to a covering space of the model. The source says that the classification of flat complete or flat tame $\mathcal{O}(n)$-geometries is unclear, so this equivalence remains open there.