This article presents a set of tools built around the Graph Rewriting computational framework which can be used to compute complex rule-based transformations on linguistic structures. Application of the graph matching mechanism for corpus exploration, error mining or quantitative typology are also given.