Technological advances in residential-scale batteries are paving the way towards self-sufficient communities to make the most use of their photovoltaic systems to support local energy consumption needs. To effectively utilize capabilities of batteries, the community can participate in the provision of short term operating reserve (STOR) services. To do so, adequate energy reserves in batteries are maintained during prescribed time windows to be utilized by electricity system operators. However, this may reduce energy sufficiency of the community. Further, the actual delivery of reserve could create distribution network congestions. To adequately understand the capability of a community to provide reserve, this work proposed a residential community energy management system formulated as a Mixed-Integer Linear Programming (MILP) model. This model aims to maximize energy sufficiency by optimal scheduling of batteries whilst considering reserve constraints. The model also maintains the aggregate power of houses within export/import limits that are defined offline using an iterative approach to ensure that the reserve provision does not breach distribution network constraints. The model is demonstrated on a residential community. The maximum committed reserve power with minimal impact on energy sufficiency is determined. Results also show that the capability of a community to provide reserve could be overestimated unless distribution network constraints are adequately considered.