How to Use the Constructor MCP in LangChain

Q: How does LangChain handle Constructor search pagination via MCP?

LangChain uses the searchpagination tool to fetch Gateway history sequentially. The agent inspects the returned token to decide if it needs to request the next page of results during a chain run.

Q: Can a LangChain agent search Constructor products?

Yes. The agent calls the searchproducts tool to search CRM records directly in the Headless Constructor.io platform. LangChain passes the query from the user input into this tool and pipes the output into your response formatter.

Q: How do I configure Constructor recommend tools in LangChain?

You register the getrecommendations tool within your LangChain agent configuration. The tool retrieves explicit cloud logging tracing vault limits, which your chain uses to filter recommendations before showing them to the user.

Q: Can you filter products by color or size?

Yes, your agent uses the searchfiltered tool to restrict product arrays to exact colors, sizes, or features. It parses the returned JSON structure to confirm the filters match the user's explicit request.

Feed Constructor search results directly into your LangChain multi-step pipelines with this MCP Server.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Constructor MCP to LangChain

Create your Vinkius account to connect Constructor to LangChain and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Constructor with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Audit Constructor account logic with LangChain

The `autocomplete` tool extracts structural properties that drive active account logic inside your multi-step LangChain pipelines. Your agent inspects these properties to determine the next logical step in a customer support flow, avoiding hardcoded routing. You get full observability through LangSmith tracing to see exactly how your agent parses these account properties. This lets you debug execution times and token usage for every single account logic check.

Route Constructor billing rules through chains

The `search_sorted` tool enumerates explicitly attached structured rules exporting active billing configurations straight to your LangChain agent. This tool enables your agent to compare active billing contracts against live search parameters during a multi-step negotiation sequence. Because LangChain handles state transitions, your agent passes these extracted billing rules directly into downstream tools without manual variable mapping. You get a clean, observable log of every rule evaluated by the model.

Extract customer bindings using the MCP Server

The `browse_category` tool provisions a JSON payload that generates hard customer bindings inside your LangChain runnable sequences. Your agent triggers this tool when a customer changes their profile preferences, instantly updating their category access. LangChain connects this payload to your vector databases or memory modules in one step. This ensures that subsequent chat turns respect the newly bound customer category rules without losing context.

Setup guide

Set up Constructor MCP in LangChain

Prerequisites

Python 3.10+ installed
langchain-mcp-adapters + langgraph packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run pip install langchain-mcp-adapters langgraph langchain-openai. The MCP adapters package converts MCP tools into native LangChain BaseTool objects.
2

Connect via HTTP transport
Use MultiServerMCPClient with "transport": "http" pointing to your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Create a ReAct agent
Pass the discovered tools to create_react_agent() from LangGraph. The agent automatically routes Constructor tool calls through the MCP protocol.
4

Run with any LLM
Swap ChatOpenAI for ChatAnthropic, ChatGoogleGenerativeAI, or any LangChain-compatible model. The MCP tools work identically across all providers.

agent.py

from langchain_mcp_adapters.client import MultiServerMCPClient
from langgraph.prebuilt import create_react_agent
from langchain_openai import ChatOpenAI

async with MultiServerMCPClient({
    "constructor-mcp": {
        "transport": "http",
        "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp",
    }
}) as client:
    tools = client.get_tools()

    agent = create_react_agent(
        ChatOpenAI(model="gpt-4o"),
        tools,
    )
    result = await agent.ainvoke({
        "messages": "List recent Constructor transactions"
    })
    print(result["messages"][-1].content)

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Constructor. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Constructor now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Constructor MCP in LangChain

LangChain uses the `search_pagination` tool to fetch Gateway history sequentially. The agent inspects the returned token to decide if it needs to request the next page of results during a chain run.

Yes. The agent calls the `search_products` tool to search CRM records directly in the Headless Constructor.io platform. LangChain passes the query from the user input into this tool and pipes the output into your response formatter.

You register the `get_recommendations` tool within your LangChain agent configuration. The tool retrieves explicit cloud logging tracing vault limits, which your chain uses to filter recommendations before showing them to the user.

Yes, your agent uses the `search_filtered` tool to restrict product arrays to exact colors, sizes, or features. It parses the returned JSON structure to confirm the filters match the user's explicit request.

This MCP Server runs inside a secure V8 isolate sandbox on Vinkius, meaning your active billing rules and customer bindings never persist on external disks. All calls to `search_sorted` occur over encrypted connections, and ephemeral runtimes wipe the memory space immediately after your LangChain agent completes its run.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript