Skip to main content

SDK

AvailableFreeAvailableIndividualAvailableTeamAvailableCustom Compare plans

The Agent SDK (airbyte_agent_sdk) is the Python interface for Airbyte Agents. With one install you get typed connectors, automatic credential handling, direct execution, and patterns for exposing connectors as tools to AI agent frameworks.

This section walks through authenticate, add a connector, and execute operations. Deeper class and method signatures live in the SDK reference.

When to use the SDK​

  • You're building a Python agent with a framework like Pydantic AI, LangChain, or FastMCP.
  • You want typed connectors and automatic credential handling in your own code.
  • You prefer in-process library calls over shelling out to a binary or making raw HTTP requests.

If your agent already supports the Model Context Protocol, the MCP server gives you zero-install access. If you prefer a shell binary, see the CLI. For non-Python backends, use the API directly.

Choose your interface​

You can use Airbyte Agents programmatically in several ways:

  • MCP server: a remote, Airbyte-hosted server. Best for AI agents that speak the Model Context Protocol (Claude, ChatGPT, Cursor, VS Code). Zero install.
  • SDK: a typed Python library. Best for Python apps, notebooks, scripts, and AI agents built with frameworks like Pydantic AI or LangChain.
  • CLI: a shell interface with composable JSON output. Best for scripts, CI jobs, and AI-agent harnesses that call command-line tools.
  • API: an HTTP API. Best for non-Python backends, custom admin flows, and languages the SDK doesn't cover.
  • Web app: a browser UI at app.airbyte.ai. Best for no-code exploration and scheduled automations.

All interfaces share the same connectors, credentials, and Context Store. See Choose how to use Airbyte Agents for a detailed comparison.

Log in and sign up​

Log in or sign up at app.airbyte.ai.

Install​

Add the SDK to a uv-managed project:

uv add airbyte-agent-sdk

To install into an existing virtual environment instead, run uv pip install airbyte-agent-sdk.

The install name uses dashes. The Python import name uses underscores: from airbyte_agent_sdk import Workspace, connect.

End-to-end example​

The example below authenticates with Airbyte and executes an operation against a GitHub connector. Before running this code, add a GitHub connector to your workspace through the web app or the REST API. The pages in this section explain each step in detail.

agent.py
import asyncio
from airbyte_agent_sdk import connect

async def main():
    # Authenticate. The SDK reads AIRBYTE_CLIENT_ID and AIRBYTE_CLIENT_SECRET
    # from the environment.

    # Execute an operation against the connector. `connect()` resolves the
    # connector by its slug within the current workspace. No ID needed
    # when the workspace has one connector of this type.
    github = connect("github")
    try:
        # Parameter names are connector- and entity-specific. Call
        # `github.list_entities()` to see what each entity accepts.
        result = await github.execute("issues", "list", params={"per_page": 10})
        for row in result.data:
            print(row)
    finally:
        await github.close()

asyncio.run(main())

Learn more​

Was this page helpful?