A Two-Dimensional Framework for AI Agent Design Patterns: Cognitive Function and Execution Topology

AI 智能体设计模式的二维框架：认知功能与执行拓扑

Abstract: 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.

摘要： 现有的基于大语言模型（LLM）的智能体架构框架通常仅从单一视角描述系统：行业指南（如 Anthropic、Google、LangChain）侧重于执行拓扑（即数据如何流动），而认知科学研究则侧重于认知功能（即智能体做什么）。仅凭其中任何一个维度都无法区分架构上截然不同的系统：相同的“编排者-执行者”（Orchestrator-Workers）拓扑可以实现“规划与执行”、“分层委派”或“对抗性验证”——这三种模式在故障模式和设计权衡方面有着本质的区别。

We propose a two-dimensional classification that combines (1) a Cognitive Function axis with seven categories (Context Engineering, 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 27 named patterns, 13 with original names.

我们提出了一种二维分类法，结合了（1）包含七个类别的认知功能轴（上下文工程、记忆、推理、行动、反思、协作、治理）和（2）包含六种结构原型的执行拓扑轴（链式、路由、并行、编排、循环、层级）。由此产生的 7x6 矩阵确定了 27 种命名模式，其中 13 种为原创命名。

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.

我们通过系统的跨轴分析证明了其正交性，详细定义了八种代表性模式，并验证了其在四个现实领域（金融借贷、法律尽职调查、网络运营、医疗分诊）中的描述覆盖范围。跨领域分析得出了五条模式选择的经验法则，这些法则支配着环境约束（时间压力、行动权限、故障成本不对称性、业务量）与架构选择之间的关系。

The framework provides a principled, framework-neutral, and model-agnostic vocabulary for AI agent architecture design.

该框架为 AI 智能体架构设计提供了一套有原则的、框架中立且与模型无关的词汇表。