# 36 terms, each with an example from a real server

GLOSSARY, read 10 Nov 2026

**The short answer.** The glossary defines 36 terms an engineer meets when wiring an agent to data tools over MCP, from tool definition and Streamable HTTP to persona scoping and monitor as code. Each term has a plain definition and an example from one of the 30 servers or the MCP specification, with its source, read 10 Nov 2026.

- [column-level lineage](/glossary/column-level-lineage/): Lineage that follows individual columns, showing which source columns feed which output column.
- [context window](/glossary/context-window/): The limit on how much text a model can read at once, including the conversation, tool definitions and every tool result.
- [Cortex Analyst](/glossary/cortex-analyst/): Snowflake's natural-language analytics service.
- [data observability incident](/glossary/data-observability-incident/): A recorded problem found by monitoring, such as late data, a failed test or a volume change.
- [Discovery API](/glossary/discovery-api/): The dbt platform API that returns metadata about a project's models, sources, exposures, tests and runs.
- [dynamic client registration](/glossary/dynamic-client-registration/): A step where an MCP client registers itself with an authorization server automatically, so nobody has to create a client ID by hand.
- [Genie space](/glossary/genie-space/): A Databricks workspace for asking questions of chosen tables in natural language, set up with curated instructions.
- [human confirmation](/glossary/human-confirmation/): A step where a person approves a tool call before it runs.
- [impact analysis](/glossary/impact-analysis/): The question of what breaks downstream if an asset changes or fails.
- [input schema](/glossary/input-schema/): A JSON Schema that states which arguments a tool accepts, their types and which are required or limited to set values.
- [local MCP server](/glossary/local-mcp-server/): An MCP server that runs on the engineer's own machine or infrastructure, usually started by the client over stdio.
- [MCP client](/glossary/mcp-client/): The part of an AI application that connects to MCP servers, lists their tools and sends the calls the model decides to make.
- [MCP server](/glossary/mcp-server/): A program that exposes tools, resources or prompts to an AI application over the Model Context Protocol.
- [mcp-remote bridge](/glossary/mcp-remote-bridge/): An open-source Node.js proxy that lets a client which only speaks stdio reach a remote MCP server.
- [metadata graph](/glossary/metadata-graph/): A connected store of data assets and their relationships: tables, columns, dashboards, owners and the links between them.
- [monitor as code](/glossary/monitor-as-code/): Defining a data monitor as configuration kept in a repository instead of clicking it together in a UI.
- [OAuth 2.1 authorization for MCP](/glossary/oauth-2-1-authorization-for-mcp/): The sign-in flow the MCP specification defines for HTTP servers.
- [persona scoping](/glossary/persona-scoping/): Limiting what an agent can see or do by the persona of the signed-in person, such as admin, member or guest.
- [personal access token](/glossary/personal-access-token/): A long-lived secret a person creates in a tool to let a script or agent act as them.
- [read-only tool](/glossary/read-only-tool/): A tool that changes nothing in the system it talks to.
- [remote MCP server](/glossary/remote-mcp-server/): An MCP server the vendor hosts and the client reaches by URL.
- [resource](/glossary/resource/): Read-only context an MCP server exposes by address, such as a file or a database schema.
- [row-level security](/glossary/row-level-security/): A rule that filters which rows a person may read in a table or model, often by role or department.
- [semantic layer query](/glossary/semantic-layer-query/): A query written against defined metrics and dimensions instead of raw tables.
- [service account](/glossary/service-account/): A non-human identity with its own role, used by unattended agents and jobs.
- [stdio transport](/glossary/stdio-transport/): A way to run an MCP server as a local subprocess.
- [Streamable HTTP](/glossary/streamable-http/): The MCP transport for servers reached over a network.
- [text-to-SQL](/glossary/text-to-sql/): A tool that turns a plain-language question into a SQL statement, usually using what it knows about the schema or project.
- [token](/glossary/token/): The unit in which models count text, roughly a word fragment.
- [tool](/glossary/tool/): A named function an MCP server offers to the model.
- [tool annotations](/glossary/tool-annotations/): Optional hints a server attaches to a tool, such as readOnlyHint, destructiveHint, idempotentHint and openWorldHint.
- [tool definition](/glossary/tool-definition/): The description of one tool that a server sends to the client: its unique name, a title, a text description and an input schema.
- [tool-definition size](/glossary/tool-definition-size/): The amount of text a server's tool definitions take: names, descriptions and input schemas together.
- [tools/list](/glossary/tools-list/): The protocol request a client sends to ask a server which tools it offers.
- [Unity Catalog function as a tool](/glossary/unity-catalog-function-as-a-tool/): A function registered in Databricks Unity Catalog and presented to an agent as an MCP tool.
- [write tool](/glossary/write-tool/): A tool that creates, changes or deletes something: a table row, a monitor, a report, a tag.

Last reviewed 10 Nov 2026. Canonical: https://datamcpservers.com/glossary/
