We report on a community effort between industry and academia to \nshape the future of graph query languages. We argue that existing \ngraph database management systems should consider supporting \na query language with two key characteristics. First, it should be \ncomposable, meaning, that graphs are the input and the output of \nqueries. Second, the graph query language should treat paths as \nfirst-class citizens. Our result is G-CORE, a powerful graph query \nlanguage design that fulfills these goals, and strikes a careful balance \nbetween path query expressivity and evaluation complexity.