Angle-constrained formation control has garnered significant attention owing to the advantage of interedge angles invariant under translation, rotation, and scaling. However, most existing approaches addressing this problem are applicable only to single-or double-integrator dynamics, which are often impractical in real-world scenarios. In this article, an angle rigidity-based adaptive formation control framework is introduced for nonlinear multiagent systems subject to mismatched uncertainties. The proposed control framework integrates a prescribed performance control approach with a recursive backstepping procedure, offering several key advantages: the capability to handle unmatched system uncertainties, the preservation of angle rigidity throughout the formation process, and the assurance that the triangulated formation shape is asymptotically achieved without risking collisions between neighboring agents. Furthermore, since the control input of each agent only requires local information related to its neighbors, which can be obtained locally from its own sensors, the proposed control method can be deployed in a communication-free environment. The effectiveness of the proposed control algorithms is validated by extensive numerical simulation.