The semantic Web will bring meaning to the Internet, making it possible for web agents to understand the information it contains. However, current trends seem to suggest that the semantic web is not likely to be adopted in the forthcoming years. In this sense, meaningful information extraction from the web becomes a handicap for web agents. In this article, we present a framework for automatic extraction of semantically-meaningful information from the current web. Separating the extraction process from the business logic of an agent enhances modularity, adaptability, and maintainability. Our approach is novel in that it combines different technologies to extract information, surf the web and automatically adapt to web changes.