In this demo paper, we propose a novel privacy-preserving data service composition scheme and an implementation thereof. Our composition system allows to execute queries over multiple data services without revealing any extra information to any of the involved services (i.e., none of involved services (and their providers) should be able to learn/infer any information about the data the other services provide beyond what is permitted). Our system is not intrusive, it assumes the services to implement locally their privacy policies before integrating their data. The system allows also to quantify the qualities of returned results relative to user’s constraints.