How to integrate Mx technologies MCP with LangChain

Trusted by
AWS
Glean
Zoom
Airtable
GET STARTED FOR FREEGET A DEMO

30 min · no commitment · see it on your stack

Mx technologies logo
LangChain logo
divider

Introduction

This guide walks you through connecting Mx technologies to LangChain using the Composio tool router. By the end, you'll have a working Mx technologies agent that can create a manual account for a user, list account numbers for a specific member, fetch rewards for a connected member through natural language commands.

This guide will help you understand how to give your LangChain agent real control over a Mx technologies account through Composio's Mx technologies MCP server.

Before we dive in, let's take a quick look at the key ideas and tools involved.

Also integrate Mx technologies with

TL;DR

Here's what you'll learn:
  • Get and set up your OpenAI and Composio API keys
  • Connect your Mx technologies project to Composio
  • Create a Tool Router MCP session for Mx technologies
  • Initialize an MCP client and retrieve Mx technologies tools
  • Build a LangChain agent that can interact with Mx technologies
  • Set up an interactive chat interface for testing

What is LangChain?

LangChain is a framework for developing applications powered by language models. It provides tools and abstractions for building agents that can reason, use tools, and maintain conversation context.

Key features include:

  • Agent Framework: Build agents that can use tools and make decisions
  • MCP Integration: Connect to external services through Model Context Protocol adapters
  • Memory Management: Maintain conversation history across interactions
  • Multi-Provider Support: Works with OpenAI, Anthropic, and other LLM providers

What is the Mx technologies MCP server, and what's possible with it?

The Mx technologies MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Mx technologies account. It provides structured and secure access to financial data aggregation and account management features, so your agent can perform actions like creating accounts, managing members, fetching financial rewards, and handling account ownership on your behalf.

  • Automated account creation and management: Let your agent create new manual accounts, partner accounts, and user SSO accounts for seamless onboarding and testing.
  • Member aggregation and connection: Instruct your agent to create members and initiate aggregation of financial products across institutions, streamlining financial data collection.
  • Rewards and incentives tracking: Have your agent fetch and aggregate member rewards data after account connections, so you never miss out on incentives.
  • Secure access to account details: Direct your agent to list account owners, retrieve account numbers by member, and access configurable widget URLs for enhanced user interactions.
  • Credential and API management: Use your agent to retrieve API credentials for audience services, streamlining authentication flows and integrations.

Supported Tools & Triggers

Tools
Disable MX User AccountDisables an MX user account, preventing background updates to accounts and transactions and restricting data access.
Create accountCreates a manual account for a specified user in MX Platform.
Retrieve Audience API CredentialsTool to retrieve Audience API credentials.
Create memberCreate a member to connect a user to a financial institution.
Create Partner AccountCreate a new user (partner account) in the MX Platform.
Fetch rewardsInitiate rewards aggregation for a specific member.
Get configurable widget URLRetrieve a configurable widget URL for embedding MX widgets (Connect, Transactions, Pulse, etc.
List Account Numbers by MemberTool to list account numbers for a specific member.
List account ownersTool to list account owners for a specific member.
List accountsTool to list all accounts for a user.
List budgetsList all budgets for a specific MX user.
List categoriesTool to list all categories for a user.
List challengesTool to list MFA challenges for a member.
List favorite institutionsList the partner's favorite financial institutions, sorted by popularity.
List goalsList all financial goals for a specific MX user.
List institution credentialsTool to list credential fields required by a given institution.
List institutionsTool to list financial institutions supported by MX.
List member accountsTool to list accounts for a specific member.
List membersList all members (financial institution connections) for a specific user.
List rewardsLists rewards and loyalty program data for a specific member.
List statements by memberRetrieve a paginated list of monthly account statements (PDF) for a specific member.
List taggingsList all taggings for a specific MX user.
List tagsList all custom tags associated with a specific MX user.
List transactionsRetrieve all financial transactions for an MX user across all their connected accounts and members.
List transactions by memberTool to list transactions for a member.
List usersList all users in the MX Platform.
Read accountTool to retrieve details for a specific account.
Read categoryTool to retrieve a default category by GUID.
Read FDX accountRetrieve details for a specific account using the FDX 4.
Read institutionTool to retrieve details for a specific institution by code.
Read transactionTool to retrieve details for a specific transaction by its GUID.
Read transaction by accountTool to retrieve details for a specific transaction by account.
Read transaction ruleTool to retrieve details for a specific transaction rule by its GUID.
Update transactionTool to update a specific transaction's description.
Update transaction rulesTool to update an existing transaction rule.
Update user taggingTool to update a tagging for an MX user.

What is the Composio tool router, and how does it fit here?

What is Composio SDK?

Composio's Composio SDK helps agents find the right tools for a task at runtime. You can plug in multiple toolkits (like Gmail, HubSpot, and GitHub), and the agent will identify the relevant app and action to complete multi-step workflows. This can reduce token usage and improve the reliability of tool calls. Read more here: Getting started with Composio SDK

The tool router generates a secure MCP URL that your agents can access to perform actions.

How the Composio SDK works

The Composio SDK follows a three-phase workflow:

  1. Discovery: Searches for tools matching your task and returns relevant toolkits with their details.
  2. Authentication: Checks for active connections. If missing, creates an auth config and returns a connection URL via Auth Link.
  3. Execution: Executes the action using the authenticated connection.

Step-by-step Guide

Prerequisites

Before starting this tutorial, make sure you have:
  • Python 3.10 or higher installed on your system
  • A Composio account with an API key
  • An OpenAI API key
  • Basic familiarity with Python and async programming

Getting API Keys for OpenAI and Composio

OpenAI API Key
  • Go to the OpenAI dashboard and create an API key. You'll need credits to use the models, or you can connect to another model provider.
  • Keep the API key safe.
Composio API Key
  • Log in to the Composio dashboard.
  • Navigate to your API settings and generate a new API key.
  • Store this key securely as you'll need it for authentication.

Install dependencies

pip install composio-langchain langchain-mcp-adapters langchain python-dotenv

Install the required packages for LangChain with MCP support.

What's happening:

  • composio-langchain provides Composio integration for LangChain
  • langchain-mcp-adapters enables MCP client connections
  • langchain is the core agent framework
  • python-dotenv loads environment variables

Set up environment variables

bash
COMPOSIO_API_KEY=your_composio_api_key_here
COMPOSIO_USER_ID=your_composio_user_id_here
OPENAI_API_KEY=your_openai_api_key_here

Create a .env file in your project root.

What's happening:

  • COMPOSIO_API_KEY authenticates your requests to Composio's API
  • COMPOSIO_USER_ID identifies the user for session management
  • OPENAI_API_KEY enables access to OpenAI's language models

Import dependencies

from langchain_mcp_adapters.client import MultiServerMCPClient
from langchain.agents import create_agent
from dotenv import load_dotenv
from composio import Composio
import asyncio
import os

load_dotenv()
What's happening:
  • We're importing LangChain's MCP adapter and Composio SDK
  • The dotenv import loads environment variables from your .env file
  • This setup prepares the foundation for connecting LangChain with Mx technologies functionality through MCP

Initialize Composio client

async def main():
    composio = Composio(api_key=os.getenv("COMPOSIO_API_KEY"))

    if not os.getenv("COMPOSIO_API_KEY"):
        raise ValueError("COMPOSIO_API_KEY is not set")
    if not os.getenv("COMPOSIO_USER_ID"):
        raise ValueError("COMPOSIO_USER_ID is not set")
What's happening:
  • We're loading the COMPOSIO_API_KEY from environment variables and validating it exists
  • Creating a Composio instance that will manage our connection to Mx technologies tools
  • Validating that COMPOSIO_USER_ID is also set before proceeding

Create a Tool Router session

# Create Tool Router session for Mx technologies
session = composio.create(
    user_id=os.getenv("COMPOSIO_USER_ID"),
    toolkits=['mx_technologies']
)

url = session.mcp.url
What's happening:
  • We're creating a Tool Router session that gives your agent access to Mx technologies tools
  • The create method takes the user ID and specifies which toolkits should be available
  • The returned session.mcp.url is the MCP server URL that your agent will use
  • This approach allows the agent to dynamically load and use Mx technologies tools as needed

Configure the agent with the MCP URL

client = MultiServerMCPClient({
    "mx_technologies-agent": {
        "transport": "streamable_http",
        "url": session.mcp.url,
        "headers": {
            "x-api-key": os.getenv("COMPOSIO_API_KEY")
        }
    }
})

tools = await client.get_tools()

agent = create_agent("gpt-5", tools)
What's happening:
  • We're creating a MultiServerMCPClient that connects to our Mx technologies MCP server via HTTP
  • The client is configured with a name and the URL from our Tool Router session
  • get_tools() retrieves all available Mx technologies tools that the agent can use
  • We're creating a LangChain agent using the GPT-5 model

Set up interactive chat interface

conversation_history = []

print("Chat started! Type 'exit' or 'quit' to end the conversation.\n")
print("Ask any Mx technologies related question or task to the agent.\n")

while True:
    user_input = input("You: ").strip()

    if user_input.lower() in ['exit', 'quit', 'bye']:
        print("\nGoodbye!")
        break

    if not user_input:
        continue

    conversation_history.append({"role": "user", "content": user_input})
    print("\nAgent is thinking...\n")

    response = await agent.ainvoke({"messages": conversation_history})
    conversation_history = response['messages']
    final_response = response['messages'][-1].content
    print(f"Agent: {final_response}\n")
What's happening:
  • We initialize an empty conversation_history list to maintain context across interactions
  • A while loop continuously accepts user input from the command line
  • When a user types a message, it's added to the conversation history and sent to the agent
  • The agent processes the request using the ainvoke() method with the full conversation history
  • Users can type 'exit', 'quit', or 'bye' to end the chat session gracefully

Run the application

if __name__ == "__main__":
    asyncio.run(main())
What's happening:
  • We call the main() function using asyncio.run() to start the application

Complete Code

Here's the complete code to get you started with Mx technologies and LangChain:

from langchain_mcp_adapters.client import MultiServerMCPClient
from langchain.agents import create_agent
from dotenv import load_dotenv
from composio import Composio
import asyncio
import os

load_dotenv()

async def main():
    composio = Composio(api_key=os.getenv("COMPOSIO_API_KEY"))
    
    if not os.getenv("COMPOSIO_API_KEY"):
        raise ValueError("COMPOSIO_API_KEY is not set")
    if not os.getenv("COMPOSIO_USER_ID"):
        raise ValueError("COMPOSIO_USER_ID is not set")
    
    session = composio.create(
        user_id=os.getenv("COMPOSIO_USER_ID"),
        toolkits=['mx_technologies']
    )

    url = session.mcp.url
    
    client = MultiServerMCPClient({
        "mx_technologies-agent": {
            "transport": "streamable_http",
            "url": url,
            "headers": {
                "x-api-key": os.getenv("COMPOSIO_API_KEY")
            }
        }
    })
    
    tools = await client.get_tools()
  
    agent = create_agent("gpt-5", tools)
    
    conversation_history = []
    
    print("Chat started! Type 'exit' or 'quit' to end the conversation.\n")
    print("Ask any Mx technologies related question or task to the agent.\n")
    
    while True:
        user_input = input("You: ").strip()
        
        if user_input.lower() in ['exit', 'quit', 'bye']:
            print("\nGoodbye!")
            break
        
        if not user_input:
            continue
        
        conversation_history.append({"role": "user", "content": user_input})
        print("\nAgent is thinking...\n")
        
        response = await agent.ainvoke({"messages": conversation_history})
        conversation_history = response['messages']
        final_response = response['messages'][-1].content
        print(f"Agent: {final_response}\n")

if __name__ == "__main__":
    asyncio.run(main())

Conclusion

You've successfully built a LangChain agent that can interact with Mx technologies through Composio's Tool Router.

Key features of this implementation:

  • Dynamic tool loading through Composio's Tool Router
  • Conversation history maintenance for context-aware responses
  • Async Python provides clean, efficient execution of agent workflows
You can extend this further by adding error handling, implementing specific business logic, or integrating additional Composio toolkits to create multi-app workflows.

How to build Mx technologies MCP Agent with another framework

OpenAI Agents SDK logo

OpenAI Agents SDK

Use Mx technologies MCP with OpenAI Agents SDK

Claude Agent SDK logo

Claude Agent SDK

Use Mx technologies MCP with Claude Agent SDK

Claude Code logo

Claude Code

Use Mx technologies MCP with Claude Code

Claude Cowork logo

Claude Cowork

Use Mx technologies MCP with Claude Cowork

Codex logo

Codex

Use Mx technologies MCP with Codex

OpenClaw logo

OpenClaw

Use Mx technologies MCP with OpenClaw

Hermes logo

Hermes

Use Mx technologies MCP with Hermes

CLI logo

CLI

Use Mx technologies MCP with CLI

Google ADK logo

Google ADK

Use Mx technologies MCP with Google ADK

Vercel AI SDK logo

Vercel AI SDK

Use Mx technologies MCP with Vercel AI SDK

Mastra AI logo

Mastra AI

Use Mx technologies MCP with Mastra AI

LlamaIndex logo

LlamaIndex

Use Mx technologies MCP with LlamaIndex

CrewAI logo

CrewAI

Use Mx technologies MCP with CrewAI

FAQ

What are the differences in Tool Router MCP and Mx technologies MCP?

With a standalone Mx technologies MCP server, the agents and LLMs can only access a fixed set of Mx technologies tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Mx technologies and many other apps based on the task at hand, all through a single MCP endpoint.

Can I use Tool Router MCP with LangChain?

Yes, you can. LangChain fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Mx technologies tools.

Can I manage the permissions and scopes for Mx technologies while using Tool Router?

Yes, absolutely. You can configure which Mx technologies scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do.

How safe is my data with Composio Tool Router?

All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Mx technologies data and credentials are handled as safely as possible.

Used by agents from

Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai

Never worry about agent reliability

We handle tool reliability, observability, and security so you never have to second-guess an agent action.

Get started for freeGet a demo↗