Existing frameworks for LLM-based agent architectures describe systems from a single perspective: industry guides (Anthropic, Google, LangChain) focus on execution topology -- how data flows -- while cognitive science surveys focus on cognitive function -- what the agent does. Neither axis alone disambiguates architecturally distinct systems: the same Orchestrator-Workers topology can implement Plan-and-Execute, Hierarchical Delegation, or Adversarial Verification -- three patterns with fundamentally different failure modes and design trade-offs. We propose a two-dimensional classification that combines (1) a Cognitive Function axis with seven categories (Perception, Memory, Reasoning, Action, Reflection, Collaboration, Governance) and (2) an Execution Topology axis with six structural archetypes (Chain, Route, Parallel, Orchestrate, Loop, Hierarchy). The resulting 7x6 matrix identifies 28 named patterns, 15 with original names. We demonstrate orthogonality through systematic cross-axis analysis, define eight representative patterns in detail, and validate descriptive coverage across four real-world domains (financial lending, legal due diligence, network operations, healthcare triage). Cross-domain analysis yields five empirical laws of pattern selection governing the relationship between environmental constraints (time pressure, action authority, failure cost asymmetry, volume) and architectural choices. The framework provides a principled, framework-neutral, and model-agnostic vocabulary for AI agent architecture design.

翻译：现有基于大语言模型的智能体架构框架多从单一视角描述系统：行业指南（如Anthropic、Google、LangChain）聚焦于执行拓扑——即数据流如何流动；而认知科学综述则聚焦于认知功能——即智能体执行何种任务。单一轴线无法区分架构迥异的系统：相同的“编排者-工作者”拓扑可同时实现“计划与执行”、“分层委派”或“对抗性验证”三种模式，而这些模式具有截然不同的失效模式与设计权衡。为此，我们提出一种二维分类体系，结合(1)包含七类范畴（感知、记忆、推理、行动、反思、协作、治理）的认知功能轴，与(2)包含六种结构原型（链式、路由、并行、编排、循环、层级）的执行拓扑轴。所构建的7×6矩阵可识别出28种命名模式（其中15种为原创命名）。我们通过系统性的跨轴分析验证其正交性，详细定义八种代表性模式，并在四个真实领域（金融借贷、法律尽调、网络运维、医疗分诊）中验证其描述覆盖能力。跨领域分析得出五项模式选择经验法则，阐明了环境约束（时间压力、行动权限、失效成本不对称性、任务规模）与架构选择之间的关联。该框架为AI智能体架构设计提供了一套原则化、框架中立且模型无关的通用词汇表。