GatewayBench puts agent routing servers through adversarial testing

A team of researchers built GatewayBench, the first shared test suite for the software servers that route agent requests to external tools. These servers sit between an AI agent and the outside world. Before this work, no shared standard existed for measuring how well they performed or how they held up under stress. The test suite covers real-world tools and includes adversarial tests, meaning tests designed to find failure modes under attack or misuse. The finding that matters most for designers is that this infrastructure layer had no agreed way to measure capability or behavior under pressure. Teams were shipping without knowing what their routing server would do under stress or manipulation. The direct design implication involves the trust a human places in an agent before handing over access to external tools. A server that can be manipulated means the human's decision to grant access was made with incomplete information. Teams building agent products should test routing server behavior and surface those results to the humans who approve agent access.