AI Agent Glossary

24 terms covering autonomous agents, tool use, orchestration, and the tools that power them.

Agent Artifact

A tangible output produced by an AI agent — documents, code, reports, or decisions that persist beyond the conversation.

concept

Agent Chaining

Connecting multiple AI agents in sequence so the output of one becomes the input of the next.

pattern

Agent Governance

Systems and rules that constrain, oversee, and validate AI agent actions — vetoes, approvals, and checks.

pattern

Agent Loop

The continuous cycle of perceive, reason, act, and reflect that drives autonomous agent behavior.

pattern

Agent Memory

Persistent storage that lets agents recall past interactions, learn from experience, and build context over time.

concept

Agent Orchestration

Coordinating multiple agents dynamically — routing tasks, managing state, and handling failures.

pattern

Agent Persona

A detailed character definition that shapes an AI agent's communication style, priorities, and decision-making.

concept

Agent Sandbox

An isolated execution environment that constrains what an AI agent can access and modify — preventing unintended side effects.

concept

Agentic Workflow

A multi-step process where an AI agent autonomously plans, executes, and iterates on tasks using tools and reasoning.

pattern

Autonomous Agent

An AI system that perceives its environment, makes decisions, and takes actions without continuous human input.

concept

ClawHub

The official skill registry and marketplace for the OpenClaw AI agent platform — "npm for AI agents."

tool

Function Calling

A provider-specific API feature that lets models output structured calls to predefined functions.

concept

LLM Routing

Directing requests to different language models based on task requirements, cost, or availability.

pattern

Model Context Protocol (MCP)

An open standard by Anthropic for connecting AI models to external tools and data sources.

concept

Model Fallback

Automatically switching to an alternative LLM when the primary model fails, is unavailable, or returns empty.

pattern

Multi-Agent System

An architecture where multiple AI agents collaborate, compete, or coordinate to accomplish tasks.

concept

Ollama

A tool for running open-source LLMs locally — download, configure, and serve models on your own hardware.

tool

OpenClaw

An open-source gateway that connects AI agents to tools and skills across messaging platforms.

tool

OpenRouter

A unified API gateway that provides access to hundreds of LLMs from multiple providers through a single endpoint.

tool

Prompt Engineering

The practice of designing inputs to LLMs to reliably produce desired outputs.

pattern

Roundtable Conversation

A structured multi-agent dialogue format where agents take turns discussing a topic, guided by an orchestrator.

pattern

Skill Marketplace

A platform where developers publish and discover modular capabilities for AI agents — the "app store" for agents.

concept

Tool Registry

A catalog of capabilities available to AI agents — tools are discovered, validated, and invoked through the registry.

pattern

Tool Use (Function Calling)

The ability of an LLM to invoke external functions — APIs, databases, code execution — as part of its reasoning.

concept