Abstract: With the rapid advancement of artificial intelligence technology, computational pathology has evolved from task-specific models toward foundation model paradigms capable of supporting multiple organs, diseases, and tasks. However, existing methods remain limited to single-image inference and are unable to fully emulate the comprehensive diagnostic processes encountered in real-world clinical scenarios. In recent years, large language models have driven the development of the agent paradigm, which, through mechanisms such as navigation, reasoning, and tool invocation, offers new approaches for modeling complex diagnostic workflows. Framed within the context of clinical pathological diagnostic processes, this article systematically reviews the key challenges and advances in computational pathology research across three stages: low-power overview, diagnostic analysis, and report generation, and discusses the current functional roles and implementation approaches of agent-related technologies at each stage. Although validation scenarios in existing studies remain limited, the core modules and methods required to construct a pathological diagnostic agent system are largely in place. Agent-related technologies are expected to facilitate the development of a truly clinically applicable intelligent assistive platform for pathological diagnosis and treatment.