This paper shows how a software toolkit enables graphic designers to make camera-based interactive environments in a short period of time without requiring experience in user interface design or machine vision. The Attentive Interaction Design Toolkit, a vision-based input toolkit, gives users an analysis of faces found in a given image stream, including facial expression, body motion, and attentive activities. This data is fed to a text file that can be easily understood by humans and programs alike. A four-day workshop demonstrated that some Flash-savvy architecture students could construct interactive spaces (e.g. Eat-Eat-Eat, TaiKer-KTV and ScreamMarket) based on a group of people's body and their head motions.