A multi-agent based prediction-system is presented in which the aim is to forecast the presence of oil slicks in a certain area of the open sea after an oil spill. In this case, the multi-agent architecture incorporates a prediction-system based on the CBR methodology, implemented in a series of interactive services, for modeling and monitoring the ocean water masses. The system’s nucleus is formed by a series of deliberative agents acting as controllers and administrators for all the implemented services. The implemented services are accessible in a distributed way, and can be accessed even from mobile devices. The proposed system uses information such as sea salinity, sea temperature, wind, currents, pressure, number and area of the slicks, etc. obtained from various satellites. The system has been trained using data obtained after the Prestige accident. The Oil Spill Multi-Agent System (OSM) has been able to accurately predict the presence of oil slicks in the north-west of the Galician coast using historical data.