The paper introduces Kernel of Partition Paths (KPP), a unified geometric representation for tree ensembles that indexes feature maps by nodes rather than splits. KPP uses a path metric to define a non-diagonal Gram matrix with a metric structure, enabling unified bounds on prediction, attribution, robustness, and generalization for regression and classification under three conditioning regimes. The robust-radius guarantee is deterministic in the KPP metric, not in raw input norms, and fast-rate refinements are posed as open problems.