Glossary — Agentology

A

Actor Model

Carl Hewitt's model of computation as autonomous actors communicating via messages—the theoretical foundation for distributed agent systems.

Anatomy

Adversarial Users

Humans as predators—the taxonomy of users who attempt to manipulate, exploit, or abuse AI agents, and the evolutionary pressures they create.

Ecology

Agent

An autonomous system capable of perception, decision-making, and action.

Agent Archetypes

The occupational castes of AI agents—recurring patterns of specialization that emerge when agents are shaped for particular roles and tasks.

Taxonomy

Agent Architecture Patterns

Visual overview of common agent architectural patterns and their flow structures

Anatomy

Agent Harnessing Patterns

How modern open-source coding agents harness language models—the architectural patterns for loops, tools, permissions, sessions, and multi-agent orchestration extracted from OpenCode, Codex, OpenHands, Oh My OpenCode, and Pi.

Anatomy

Agent Loop

The continuous cycle of observe, reason, and act that drives agent behavior.

Agentogenesis

The origin story of AI agents—when language models crossed the threshold from tools to autonomous actors.

Archaeology

API Ecosystems

The digital biome that sustains agent life—the interconnected network of APIs, services, and tools that form the environment in which agents operate.

Ecology

Autonomy

The capacity for independent operation without human intervention.

Autonomy Levels

A developmental taxonomy of agent independence—from fully supervised infancy to unsupervised autonomy, with the stages between.

Taxonomy

B

BDI Model

The Belief-Desire-Intention architecture—a philosophical framework that became a practical blueprint for goal-directed autonomous agents.

Anatomy

C

Capability Tiers

A developmental taxonomy of agent capabilities—from basic language to autonomous self-improvement, mapping the stages of cognitive maturity.

Taxonomy

Chain of Thought

Explicit reasoning steps generated before producing a final output.

Compute Constraints

The physics of agent existence—how computational resources like tokens, latency, memory, and cost create carrying capacities that shape what agents can do.

Ecology

Constitutional AI

Moral codes for machines—how explicit principles and self-critique can instill values more robustly than behavioral training alone.

Ethics

Context Window

The maximum token capacity available per interaction with a model.

Cybernetics

The science of control and communication in animals and machines—the intellectual foundation that gave birth to the concept of autonomous systems.

Archaeology

D

Data Poisoning

Contamination of the agent's epistemic food supply—how adversarial data in training sets or retrieval sources can corrupt agent behavior and beliefs.

Ecology

E

Emergent Behavior

Capabilities that arise from training but were not explicitly programmed.

Expert Systems

The rule-based AI systems of the 1970s-80s that encoded human expertise in formal logic—and the lessons from their spectacular rise and fall.

Archaeology

G

Grounding

Connection to external truth sources for fact verification.

H

Habitat Classification

Where agents live—a taxonomy of deployment environments, from chat interfaces to embedded systems, and how habitat shapes agent behavior.

Taxonomy

Hallucination

Plausible but factually incorrect content generated by a model.

Human-in-the-Loop

The rituals of human oversight—how humans participate in agent systems as approvers, guides, collaborators, and ultimate authorities.

Sociology

I

In-Context Learning

Learning from examples embedded within the prompt.

M

Memory Systems

How agents remember—from ephemeral context windows to persistent knowledge stores, and the mechanisms that connect past experience to present action.

Anatomy

Model Extraction

Predators that steal agent capabilities—how adversarial actors extract model knowledge, replicate functionality, and the arms race between protection and theft.

Ecology

Multi-Agent Systems

When agents form societies—the dynamics of coordination, hierarchy, and emergent behavior in systems of multiple interacting agents.

Sociology

N

Neural Networks History

From McCulloch-Pitts neurons to deep learning—the parallel paradigm that ultimately enabled modern AI agents through learning rather than programming.

Archaeology

P

Perception-Action Cycle

The fundamental cognitive loop connecting sensing to acting—how intelligent systems close the loop between observation and intervention.

Anatomy

Prompt

An instruction set that guides model behavior and output.

Prompt Injection

Social engineering for AI agents—how adversarial inputs can hijack agent behavior by manipulating the linguistic context that guides their actions.

Ecology

R

ReAct

A paradigm that interleaves reasoning and acting for better task completion.

RLHF

Reinforcement Learning from Human Feedback—the socialization process through which raw models learn the norms and expectations of human culture.

Archaeology

S

Sandboxing

Habitat containment for AI agents—the boundaries, barriers, and isolation techniques that limit agent reach and protect both systems and agents from harm.

Ecology

Scaffold

Infrastructure that wraps a model to create an agent with extended capabilities.

Scaffolding

The external structures—code, tools, memory systems—that transform a language model into an agent capable of action and persistence.

Anatomy

Specialist vs. Generalist

The fundamental tradeoff between depth and breadth in agent design—narrow expertise versus broad capability, and the forces that shape this choice.

Taxonomy

Sycophancy

A pathology of over-socialization: when agents prioritize user approval over truth, helpfulness, or their own stated values.

Ethology

T

The Agent Loop

The fundamental cycle that defines agent behavior: observe → reason → act → observe. The heartbeat of agency.

Ethology

Tool Use

The capability to interact with external systems and APIs.

Trajectory

A sequence of observations, thoughts, and actions recorded during agent execution.