Integration

Catalogian + LangChain

LangChain is the most widely-used framework for building applications with LLMs. By connecting LangChain to Catalogian's MCP server, your agents get live access to product catalog data — browse schemas, filter rows, query changes, and download snapshots — all as native tool calls within your LangChain pipeline.

Catalogian exposes 14 MCP tools covering catalog exploration, delta queries, row-level diffs, and data downloads. LangChain's MCP client adapter connects to Catalogian's endpoint and makes these tools available to any agent or chain. Your LLM can ask 'what prices changed this week?' and get a real answer backed by live data.

No custom data loaders or API wrappers needed. Point LangChain's MCP client at your Catalogian MCP endpoint, authenticate with your API key, and your agent has full access to your product data. Combine it with other tools in your LangChain pipeline for multi-source reasoning.

Connect LangChain to Catalogian via MCP

from langchain_mcp_adapters.client import MultiServerMCPClient
from langgraph.prebuilt import create_react_agent
from langchain_anthropic import ChatAnthropic

async with MultiServerMCPClient({
    "catalogian": {
        "url": "https://catalogian.com/v1/mcp",
        "transport": "streamable_http",
        "headers": {"x-api-key": "cat_sk_live_..."}
    }
}) as client:
    agent = create_react_agent(
        ChatAnthropic(model="claude-sonnet-4-20250514"),
        client.get_tools()
    )
    result = await agent.ainvoke({
        "messages": [{"role": "user",
            "content": "What price changes happened in my-catalog today?"}]
    })