Context Layer Glossary
This glossary defines terms used in Fivetran Context Services and its documentation.
For definitions of general Fivetran terms, see the Fivetran glossary.
Context
Context is the knowledge specific to your organization that AI agents and assistants draw on to answer questions accurately.
Context Layer
Context Layer is the Fivetran product that creates a unified context layer for your organization by connecting your documentation sources, systems of record, and analytical tools. A context layer is the structured representation of your business definitions, operational data, and analytics metadata. Once built, any AI tool or application can query that context layer through a single endpoint.
Learn more in our Context Layer documentation.
Agent
An agent is an AI system that uses your context layer to answer questions or take actions on your organization's behalf. Popular agents include Claude, Cursor, ChatGPT, Codex, and Gemini.
Model Context Protocol (MCP)
The Model Context Protocol (MCP) is an open protocol that enables AI coding and chat tools, custom agents, and applications to query external context sources, such as a context layer, in a consistent and predictable way.
Agent–User Interaction Protocol (AG-UI)
Agent–User Interaction Protocol (AG-UI) is a protocol that enables end users to interact directly with AI agents through the applications they already use, such as Slack and Zendesk. AG-UI connects those applications to agents that have access to your context layer.
Agent Context MCP
Agent Context MCP is the Fivetran service that exposes your organization's context layer to AI agents, coding assistants, and AG-UI applications through the MCP. Learn more in the Agent Context MCP documentation.
Agents Schema
Agents Schema is an open-source standard that designates a schema in your data warehouse as the shared context layer for AI agents. Learn more in our Context Layer documentation.