We are interested in managing documents in virtual libraries specialised in the engineering science field (ESF). Our goal is to design an intelligent tutoring system for personalised information retrieval depending on the user's interests. The proposed system is specially designed for people having sight deficiency. These specific persons use slow devices in order to access textual information, so they have a critical need for systems able to retrieve relevant information quickly. For these reasons, we need intelligent systems to delegate tasks and to make decisions. Our goal is to define a multiagent system for retrieving and filtering documents in these specific virtual libraries. We aim at improving answer quality by personalising the retrieval and by performing user's profile management. The multiagent solution proposed in this work is based on co-operative and adaptive information filtering and retrieval.