Skip to main content

Agent Engine MCP server

Airbyte's Agent Engine MCP server lets AI agents interact with your data sources through the Model Context Protocol (MCP). Instead of writing code to fetch data, you configure a connector and register it with your agent. The agent then calls the connector's operations automatically based on your natural language prompts.

What the MCP server does​

The MCP server sits between your agent and Airbyte's agent connectors. It exposes each connector's operations as MCP tools that AI assistants can discover and call. When you prompt your agent, like show me the 10 most recent Gong calls, the agent calls the MCP server, which calls the connector, which calls the API.

The server handles:

  • Discovering available entities and actions from the connector

  • Executing operations and returning structured results

  • Truncating large text fields to reduce token usage

  • Field selection and exclusion to control what data the API returns

  • Downloading binary content like call recordings

When to use the MCP server​

The MCP server is the fastest way to give an AI coding tool access to your data. It requires no code. You configure a YAML file, register the server, and start prompting.

How it works​

You configure each server instance with a YAML file that specifies which connector to use and how to authenticate. You can run a single connector per server, or use an aggregate configuration to run multiple connectors in one server.

The server supports two execution modes:

  • Open source mode: The server installs a connector package (from PyPI, a git URL, or a local path) and calls the third-party API directly using your credentials.

  • Hosted mode: The server calls the Airbyte Cloud API, which manages credentials and provides additional capabilities like indexed search through the context store.

A command line tool called adp manages the full lifecycle:

  1. Discover connectors with adp connectors list-oss or adp connectors list-cloud.

  2. Configure a connector with adp connectors configure, which generates a YAML config file.

  3. Register the server with your AI tool using adp mcp add-to.

  4. Use the server by prompting your AI tool with natural language questions.

Get started​

Follow the Agent Engine MCP server tutorial to install the MCP server, configure a connector, and register it with your AI tool. You can be up and running in about 5 minutes.

Was this page helpful?