Contemporary information systems employ REST- or Broker-based architectures, using standards like OpenAPI for information access and management and user-friendly interfaces like dashboards, graphs, or tables, allowing users to access and interact with them efficiently and practically. At the same time, Conversational Assistants (CAs) are transforming the entire user interaction with data and information, offering more intuitive and natural interfaces and data exchange. However, their adoption is delayed due to several challenges, including the lack of standardized automation tools, complex dialogue flow design, and a steep learning curve of the CA domain. To this end, we propose a low-code pipeline that converts an OpenAPI specification into ready-to-deploy CAs, embedded into Locsys, a rapid, low-code application development platform. Using Model-Driven Engineering and low-code approaches, the proposed pipeline employs and extends dFlow, a Domain-Specific Language for CA development, by automating the OpenAPI-to-dFlow transformation process leveraging Large Language Models, and ultimately generating complete CA scenarios from OpenAPI models. The proposed solution is validated on its technical soundness using OpenAPI services of varying complexity, measuring the accuracy and consistency of the results and the productivity gain compared to traditional CA development processes. The results show that the presented low-code pipeline enables the conversationalization of information systems using OpenAPI specifications, mitigating domain expertise and knowledge barriers, reducing development time from hours to minutes, and thus providing a valid alternative to existing information system interfaces.