Agent Reasoning Loop

Agent Memory

The storage mechanism enabling AI agents to retain and recall information from previous interactions and experiences.

Agent Loop

The iterative cycle of perception, reasoning, planning, and action execution that drives autonomous agent behaviour.

ReAct Framework

Reasoning and Acting — a framework where language model agents alternate between reasoning traces and action execution.

Agent Reflection

The ability of an AI agent to evaluate its own outputs and reasoning, identifying errors and improving responses.

Task Decomposition

Breaking down complex tasks into smaller, manageable subtasks that can be distributed among AI agents.

Agent Benchmarking

Standardised evaluation of AI agent capabilities across tasks like tool use, planning, reasoning, and task completion.

Agent Memory Bank

A persistent knowledge store that enables AI agents to accumulate and recall information across sessions, supporting long-term learning and personalised interactions.

Plan-and-Execute Pattern

An agentic architecture where a planning module decomposes goals into ordered tasks and a separate executor carries them out, enabling complex multi-step problem solving.

Agentic RAG

An advanced retrieval-augmented generation pattern where an agent dynamically decides what information to retrieve, from which sources, and how to refine queries iteratively.

An AI agent that selects actions to maximise a utility function representing the desirability of different outcomes.

A mechanism allowing language models to invoke external functions or APIs based on natural language instructions.

Safety constraints and boundaries that limit agent behaviour to prevent harmful, unintended, or out-of-scope actions.

Belief-Desire-Intention — an agent architecture where agents reason about beliefs, desires, and intentions to decide actions.

The accumulated information, history, and environmental state that informs an AI agent's decision-making.

The process by which AI agents reach agreements through offers, counteroffers, and compromise strategies.

The sequential composition of multiple AI agents where each agent's output becomes the input for the next, creating automated pipelines for complex multi-stage processes.

A theoretical framework that models the structure and processes of the human mind for building intelligent agents.