Agentic Artificial Intelligence (AI): Architectures, Taxonomies, and Evaluation of Large Language Model Agents

Агентный искусственный интеллект (ИИ): архитектуры, таксономии и оценка агентов на основе больших языковых моделей
Rajkumar Buyya, Arunkumar .V, Gangadharan G R
2026-01-18

Agent architecturesAgentic AILLM agentsModel Context ProtocolPrompt injection
Artificial Intelligence is moving from models that only generate text to Agentic AI, where systems behave as autonomous entities that can perceive, reason, plan, and act. Large Language Models (LLMs) are no longer used only as passive knowledge engines but as cognitive controllers that combine memory, tool use, and feedback from their environment to pursue extended goals. This shift already supports the automation of complex workflows in software engineering, scientific discovery, and web navigation, yet the variety of emerging designs, from simple single loop agents to hierarchical multi agent systems, makes the landscape hard to navigate. In this paper, we investigate architectures and propose a unified taxonomy that breaks agents into Perception, Brain, Planning, Action, Tool Use, and Collaboration. We use this lens to describe the move from linear reasoning procedures to native inference time reasoning models, and the transition from fixed API calls to open standards like the Model Context Protocol (MCP) and Native Computer Use. We also group the environments in which these agents operate, including digital operating systems, embodied robotics, and other specialized domains, and we review current evaluation practices. Finally, we highlight open challenges, such as hallucination in action, infinite loops, and prompt injection, and outline future research directions toward more robust and reliable autonomous systems.
1
Agent architectures range from simple single-loop agents to hierarchical multi-agent systems, reflecting increasing structural diversity and complexity.
2
Agent evaluation spans digital operating systems, embodied robotics, and specialized domains, while major unresolved risks include action hallucination, infinite loops, and prompt injection.
3
Agentic AI extends LLMs from passive text generation to autonomous systems that perceive, reason, plan, act, use tools, and pursue extended goals.
4
The field is transitioning from linear reasoning and fixed API calls toward inference-time reasoning models, Model Context Protocol, and native computer-use capabilities.
5
The paper proposes a unified agent taxonomy organized around Perception, Brain, Planning, Action, Tool Use, and Collaboration.

Agentic AI systems based on Large Language Model agents

Architectures, taxonomies, operating environments, evaluation practices, and reliability challenges of autonomous LLM agents

Publication Details
Publication Date
2026-01-18
Journal
Publisher
ISSN
Cited by
37
Access Type
Author Information
Authors
Rajkumar Buyya
Arunkumar .V
Gangadharan G R
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%