The paper describes a management model where a service provider (SP) maintains multiple protocol modules to exercise the infrastructure resources (e.g., bandwidth, storage, servers) under various environment conditions. Each protocol depicts a distinctive work-flow among the networked entities (i.e., message exchanges and task coordination), incurring a certain amount of resource use. Different levels of performance optimality and fault resilience are thus achieved by various protocols in any given operating region of the network infrastructure and the environment. At run-time, the SP selects one of the protocol modules that meet the application's Quality of Service (QoS) specs against the prevailing operating conditions. A single shoe does not fit all sizes !! Our model allows a dynamic switching from one protocol module to another based on the changing environment conditions. A protocol configuration that is responsive to the dynamically changing conditions manifests as a highly agile and adaptive QoS offering. An enterprise service, for instance, may change its internal work-flow processes & procedures to quickly adapt to the changing business environment. The paper describes case studies of enterprise systems and content networks to exemplify the protocol switching as a foundation to build agile network services.