In this paper, we present the software architecture of a new generation of advisory systems using Intelligent Agent and Semantic Web technologies. Software agents act as clients and advisors to implement negotiation processes in a consultancy situation. The domain knowledge is modeled with the OWL ontology language. Using an inference engine the agents reason on base of their knowledge to make decisions or proposals. The agent knowledge consists of private data and publicly accessible knowledge spread over different web sites. As in a real consultancy an agent only reveals sensitive private data, if it is indispensable for finding a solution. Depending on the actual consultancy situation each agent dynamically expands its knowledge base by accessing OWL knowledge sources from the Internet.