Interfaces
Airbyte Agents supports four interfaces for working with your data and your agents. Choose the interface that best fits how you want to build and run agents.
- Web app: The Airbyte Agents web app at app.airbyte.ai. Talk to an Airbyte-hosted agent in Chats, or define Automations that run manually or on a schedule. Best when you want Airbyte itself to be your agent, with no code required.
- MCP server: A remote, Airbyte-hosted Model Context Protocol server that connects MCP-capable agents (like ChatGPT, Claude, and Cursor) to your data. Best for conversational agents that use off-the-shelf clients.
- SDK: Python SDK for building, authenticating, and executing agent connectors directly in your Python applications. Best for Python-based agents you build and host yourself.
- API: HTTP API for managing connectors, tokens, and executing operations from any language or backend service. Best for non-Python backends, custom admin flows, and embedding the authentication module in your app.
Web app
Talk to an Airbyte-hosted agent in Chats, or define Automations that run on a schedule or webhook.
SDK
Build, authenticate, and execute agent connectors directly in your Python applications.
API
Manage connectors, tokens, and operations from any language or backend service.
MCP
The Agent MCP connects your AI agent to your data through the Model Context Protocol (MCP). It gives your agent authenticated access to the platforms you use every day, like your CRM, support desk, analytics tools, and more, so your agent can read and write data on your behalf. See Connectors for a list of available connectors.