Currently,　agent-based　computing　is　an　active　research　area,　and　great　efforts　have　been　made　towards　the　agent-oriented　programming　both　from　a　theoretical　and　practical　view.　However,　most　of　them　assume　that　there　is　no　uncertainty　in　agents＇　mental　state　and　their　environment.　In　other　words,　under　this　assumption　agent　developers　are　just　allowed　to　specify　how　his　agent　acts　when　the　agent　is　100%　sure　about　what　is　true/false.　In　this　paper,　this　unrealistic　assumption　is　removed　and　a　new　agent-oriented　probabilistic　logic　programming　language　is　proposed,　which　can　deal　with　uncertain　information　about　the　world.　The　programming　language　is　based　on　a　combination　of　features　of　probabilistic　logic　programming　and　imperative　programming.