Service creation, or service engineering, is widely studied in the telecommunication community (e.g., ITU-T, ETSI, TINA-C, RACE, ACTS or EURESCOM). All these studies adopt a common approach that is the definition of an architecture and of a methodological framework with its support. Thus the IN and TINA architectures have been defined. The methodological aspect is described by the service creation environment (SCE) concept. This enables one to unify the process of the service creation by defining a role model, a service life cycle model and a set of methods and tools that support the activities of all the roles. Concurrently with these studies, the development of applications in the Internet shows a new way to design telecommunication services based on the agent paradigm. An agent is a computational entity which acts on behalf of other entities in an autonomous way, performs its actions with some level of proactivity and/or reactiveness and exhibits some level in the key attributes of learning, cooperation and mobility. It presents adaptation and interaction capabilities that provide the flexibility required by the service creation process. In addition to agent technology, we use formal methods in order to ensure the quality, i.e. robustness and reliability of the service. This approach is based on a software engineering process and is in conformance with the reference model of open distributed processing (RM-ODP) standards.