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

TL;DR

The paper introduces a two-dimensional classification framework for AI agent architectures, combining cognitive functions and execution topologies to systematically identify and analyze 27 distinct design patterns.

Contribution

It proposes a novel 7x6 matrix model that categorizes AI agent patterns, demonstrating its applicability across multiple real-world domains and deriving empirical laws of pattern selection.

Findings

Identifies 27 AI agent design patterns with 13 newly named.

Validates the framework across four diverse real-world domains.

Derives five empirical laws relating environmental constraints to architectural choices.

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. 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. 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.