In declarative programming languages based on the constraint programming paradigm, computations can be viewed as deductions and are enhanced with the use of constraint solvers. However, admissible constraints are restricted to formulae handled by solvers and thus, declarativity may be jeopardized. We present a domain-independent scheme for extending constraint solvers with new function symbols. This mechanism, called Solex, consists of a collaboration of elementary solvers. They add and deduce information related to constraints involving new functions, complete the computation domain and purify constraints. Some extensions of computation domains have already been studied to demonstrate the broad scope of Solex potential applications.