A personal external service agent is an AI entity authorized and continuously managed by a specific individual, capable of receiving client requests, delivering work, and offering ongoing support using that person’s knowledge, experience, professional methods, and service boundaries. The concept lets individuals serve more clients with less direct involvement and generate economic value at scale. To evaluate such platforms systematically, we introduce the DelegateEval framework. It spans eight dimensions: ease of creation, service discovery and matching, delivery quality, expertise inheritance, service continuity, learning and improvement, reliability and governance, efficiency and value. Each dimension checks whether a platform enables people from diverse backgrounds to turn expertise into verifiable ongoing services, to improve capabilities through experience, and to remain economically feasible. We present the framework description and an experimental design outline; complete datasets, executable environments, a unified scoring protocol, and cross‑model results are still to be built and empirically validated.
Review: The framework offers a comprehensive lens for assessing personal AI service platforms, highlighting end‑to‑end factors from creation to value generation, and sets a clear roadmap for future implementation.