Critical infrastructures, such as energy systems, transportation and telecommunications networks, are essential for the safety and socioeconomic stability of a society. Critical infrastructures are often tightly coupled and interdependent on each other, and the topology of the interdependencies between different systems, also referred to as the coupling interface, plays a key role in terms of their performance and resilience against failures. In case of failures due natural events or deliberate attacks, the design of the coupling interface can strongly impact the systems performance. However, in the existing literature, the issue of the coupling interface design is often addressed approximately. In this work, we propose an optimization-based mathematical approach for designing coupling interfaces between interdependent critical infrastructures under external attacks. Given a set of possible attack scenarios, the proposed approach allows designing a coupling interface such that the interdependent infrastructures are robust against the worst-case realization of performance losses. Using as a case-study interdependent power and gas networks, we show that the proposed method outperforms existing solutions based on network metrics.