In this paper we present a brief overview of the ΦLOG project, aimed at the development of a domain specific framework for the rapid prototyping of applications in evolutionary biology. This includes the development of a domain specific language, called ΦLOG, and an agent-based implementation for the monitoring and execution of ΦLOG’s programs. A ΦLOG program – representing an intended application from an evolutionary biologist – is a specification of what to do to achieve her/his goal. The execution and monitoring component of our system will automatically figure out how to do it. We achieve that by viewing the available bioinformatic tools and data repositories as web services and casting the problem of execution of a sequence of bioinformatic services (possibly with loops, branches, and conditionals, specified by biologists) as the web services composition problem.