Developer Quickstart

This section is for developers who want to build agents that use Airbyte connectors. Pick a path based on how you prefer to work.

Tutorials

Step-by-step guides that take you from an empty project to a working agent in approximately 15 minutes. Each tutorial uses a different Python framework, but all of them follow the same pattern: create a project, install the Airbyte Agent SDK, connect to a data source, and execute operations.

Pick the framework you use:

• Pydantic AI — Build an agent with Pydantic AI and the Airbyte Agent SDK.
• LangChain — Build an agent with LangChain, LangGraph, and the Airbyte Agent SDK.
• FastMCP — Build an MCP server with FastMCP and the Airbyte Agent SDK.

Skills

Skills are pre-packaged instructions you install into a coding agent so it can generate working code against Airbyte connectors. Instead of writing the integration yourself, you install a skill and prompt the agent to build with Airbyte.

Pick the coding agent you use:

• Claude Code — Install Airbyte skills into Claude Code as a plugin or through skills.sh.
• Codex — Install Airbyte skills into OpenAI Codex through skills.sh or a manual symlink.
• Lovable — Paste the Airbyte skill into a Lovable build prompt to generate full-stack apps with Airbyte connectors.

Before you start

Both paths share a few common requirements.

• An Airbyte Agents account. Sign up for free at app.airbyte.ai.
• API credentials. Copy your AIRBYTE_CLIENT_ID and AIRBYTE_CLIENT_SECRET from the Profile page in the Airbyte Agents web app. See Manage your user profile for details.
• A connector. Add at least one connector to your workspace from the Connectors page in the web app. The tutorials use GitHub, but any connector works.
• Python 3.13+ and uv (for tutorials). Skills have their own prerequisites listed on each skill page.

FastMCP

In this tutorial, you'll create a new Python project with uv, build a FastMCP server that exposes one of Airbyte's agent connectors as an MCP tool, and use it to query GitHub data from any MCP-compatible agent. This tutorial uses GitHub, but if you don't have a GitHub account you can swap in any other agent connector and perform different operations.

LangChain

In this tutorial, you'll create a new Python project with uv, add a LangChain agent, equip it with one of Airbyte's agent connectors, and use natural language to explore your data. This tutorial uses GitHub, but if you don't have a GitHub account you can swap in any other agent connector and perform different operations.

Pydantic AI

In this tutorial, you'll create a new Python project with uv, add a Pydantic AI agent, equip it with one of Airbyte's agent connectors, and use natural language to explore your data. This tutorial uses GitHub, but if you don't have a GitHub account you can swap in any other agent connector and perform different operations.

Skills

Install pre-packaged instructions into a coding agent so it can generate working code against Airbyte connectors.