Firstly, this paper provides a definition of agents including properties and characteristics, and discusses the similarities and differences between agent technology and traditional artificial intelligence, expert systems, as well as objects. The paradigm, structure, and types of agents, i.e. cognitive and reactive agents, are also discussed. Secondly, this paper discusses the development of Multi-Agent Systems (MASs) and the fundamental issues to be considered when designing a MAS, followed by the delineations of possible benefits drawn when applying the concept of MASs. Thirdly, this paper discusses Agent Communication Languages (ACLs) introduced by both the US-based DARPA-funded Knowledge Sharing Effort (KSE), which developed Knowledge Query Manipulation Language (KQML) and Knowledge Interchange Format (KIF), and the Foundation for Intelligent Physical Agents (FIPA). In the same section, agent development tools are also listed, some of which are commercially available and some are still in the research stage.