Fuel cells are electrochemical devices that convert the chemical energy of a gaseous fuel directly into electricity. They are widely regarded as potential future stationary and mobile power sources. The response of a fuel-cell system depends on the air and hydrogen feed, flow and pressure regulation, and heat and water management. In this article, the study is concentrated on the air subsystem that feeds the fuel-cell cathode with oxygen and, in particular, on the problem of providing tuning rules for these controllers ensuring stability of the overall system. Proceeding from a reduced order non-linear model, that preserves the main features of the (by-now classical) ninth order model, we suggest a natural decomposition into interconnected subsystems where one of them is strictly passive, hence finite L2-gain stable, and the other one depends on the controller parameters. The proposed tuning methodology consists then on enforcing the required input-output property of the feedback loop, either passivity or a suitable L2-gain. For this end, the feedback operator is linearised, then robust Kharitonov-based positive (or bounded) realness conditions are imposed to determine the allowable ranges for the controller gains. We illustrate the methodology with a classical cascaded loop-controller structure with an inner loop feedback linearising controller and an outer loop PI regulator. Simulation results are presented to illustrate the conservativeness of the analysis as well as the performance improvement obtained with a suitable tuning.