The User-PC computing system (UPC) has been devised to provide a very low-cost distributed computing platform to members of a group, using idling resources of their personal computers (PCs). Based on the master-worker model, the master PC accepts jobs from users and assigns them to available worker PCs. Unfortunately, an efficient job assignment method has not been implemented yet. In this paper, we propose a static job scheduling algorithm considering the CPU core utilization, for the UPC system. Given a set of independent jobs, this two-stage heuristic algorithm finds an assigned worker for each job in order to minimize the makespan. To efficiently utilize CPU cores in worker PCs, the first stage groups workers and jobs into several classes according to the number of available cores or threads. It then greedily sets up job-worker assignments in each class independently. The second stage improves the assignments with a local search method by randomly moving job-worker assignments between different classes. For evaluation, we conducted experiments using six worker PCs and up to 27 jobs. Our algorithm could reduce the makespan by up to 60% compared to three baseline scheduling algorithms. However, its performance gradually decreases as the number of jobs significantly increases.