We extend the principal/one-agent model with countervailing incentives to a framework in which the principal deals with two agents behaving non-cooperatively and protected by limited liability. Focusing on the two-type case, we show that, beside the situation in which first best is effected even...