Using the SDK

The Alation AI Agent SDK enables AI agents to access and leverage metadata from the Alation Data Catalog. This SDK empowers developers to build sophisticated AI agents that can search, retrieve, and reason about your data catalog.

When to use the SDK

The SDK is primarily for local mode integrations where you want to:

• Build custom agents in Python using frameworks like LangChain
• Run a local MCP server with full control over configuration
• Integrate Alation directly into your Python applications
• Customize and extend the SDK’s functionality

If you’re using remote MCP (covered in our MCP Client Setup guides), you don’t need to install the SDK. The remote MCP server at https://<YOUR_INSTANCE>.alationcloud.com/ai/mcp/ handles everything for you.

Components

The SDK consists of three main packages:

Core SDK (alation-ai-agent-sdk)

The foundation that handles authentication, API calls, and provides core tools for interacting with the Alation catalog.

LangChain integration (alation-ai-agent-langchain)

Adapters for building sophisticated multi-agent workflows using the LangChain framework.

MCP integration (alation-ai-agent-mcp)

A local MCP server implementation that exposes Alation capabilities to any MCP client.

Quick start

pip install uv

# Install the core SDK
uv pip install alation-ai-agent-sdk==1.0.0rc2

# Install LangChain integration (optional)
uv pip install alation-ai-agent-langchain==1.0.0rc2

# Install the MCP integration (optional)
uv pip install alation-ai-agent-mcp==1.0.0rc2

Basic usage

from alation_ai_agent_sdk import AlationAPI, ServiceAccountAuthParams

# Initialize with service account authentication
auth_params = ServiceAccountAuthParams(
    client_id="your_client_id",
    client_secret="your_client_secret"
)

alation_api = AlationAPI(
    base_url="https://your-instance.alationcloud.com",
    auth_method="service_account",
    auth_params=auth_params
)

Learn more

For comprehensive documentation, examples, and guides:

Alation AI Agent SDK Complete SDK documentation, guides, and examples on GitHub.