How to integrate Synthflow ai MCP with Vercel AI SDK v6

Trusted by
AWS
Glean
Zoom
Airtable

30 min · no commitment · see it on your stack

Synthflow ai logo
Vercel AI SDK logo
divider

Introduction

This guide walks you through connecting Synthflow ai to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Synthflow ai agent that can create a new ai assistant for customer support, list all current voice assistants in your account, fetch details for team 'sales outreach' through natural language commands.

This guide will help you understand how to give your Vercel AI SDK agent real control over a Synthflow ai account through Composio's Synthflow ai MCP server.

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

Also integrate Synthflow ai with

TL;DR

Here's what you'll learn:
  • How to set up and configure a Vercel AI SDK agent with Synthflow ai integration
  • Using Composio's Tool Router to dynamically load and access Synthflow ai tools
  • Creating an MCP client connection using HTTP transport
  • Building an interactive CLI chat interface with conversation history management
  • Handling tool calls and results within the Vercel AI SDK framework

What is Vercel AI SDK?

The Vercel AI SDK is a TypeScript library for building AI-powered applications. It provides tools for creating agents that can use external services and maintain conversation state.

Key features include:

  • streamText: Core function for streaming responses with real-time tool support
  • MCP Client: Built-in support for Model Context Protocol via @ai-sdk/mcp
  • Step Counting: Control multi-step tool execution with stopWhen: stepCountIs()
  • OpenAI Provider: Native integration with OpenAI models

What is the Synthflow ai MCP server, and what's possible with it?

The Synthflow ai MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Synthflow ai account. It provides structured and secure access to your voice automation tools, so your agent can perform actions like managing voice assistants, handling teams, retrieving phone numbers, and automating call center operations on your behalf.

  • AI assistant management: Create, list, update, or delete AI-powered voice assistants to tailor customer interactions and automate call flows as needed.
  • Team creation and configuration: Set up new teams, modify existing ones, or remove teams to optimize your call center's routing and operational structure.
  • Knowledge base integration: Retrieve and manage knowledge base details to ensure your assistants have accurate, up-to-date information for conversations.
  • Phone number administration: Fetch and organize phone numbers linked to your workspace, making it easy to assign or reassign numbers for inbound and outbound campaigns.
  • Comprehensive assistant and team insights: Access detailed metadata and configuration for both assistants and teams, streamlining oversight and decision-making for your AI-powered operations.

Supported Tools & Triggers

Tools
Add document to knowledge base sourceTool to add a document to a knowledge base source.
Attach Actions to AgentTool to attach one or more actions to an agent.
Attach contact to memory storeTool to attach a contact to a memory store.
Attach knowledge base to agentTool to attach a knowledge base to an agent.
Attach memory store to agentTool to attach a memory store to an agent.
Create ActionTool to create a new action in Synthflow AI.
Create AssistantTool to create a new assistant.
Create a contactTool to create a new contact in Synthflow AI.
Create knowledge baseTool to create a new knowledge base and return its ID.
Create memory storeTool to create a new memory store.
Create Phone BookTool to create a new phone book.
Create phone book entryTool to create a phone book entry.
Create Simulation CaseTool to create a new simulation case.
Create a simulation scenarioTool to create a new simulation scenario.
Create a new simulation suiteTool to create a new simulation suite attached to a specific agent.
Create a new teamTool to create a new team.
Delete an actionTool to delete an existing action.
Delete an assistantTool to delete an existing AI assistant.
Delete a chat sessionTool to delete a chat session.
Delete a contactTool to delete an existing contact.
Delete knowledge baseTool to delete an existing knowledge base.
Delete a knowledge base sourceTool to delete a source from a knowledge base.
Delete a memory storeTool to delete a memory store.
Delete a phone bookTool to delete an existing phone book.
Delete a phone book entryTool to delete a phone book entry.
Delete a simulation caseTool to delete a simulation case by ID.
Delete a simulation scenarioTool to delete an existing simulation scenario.
Delete a simulation suiteTool to delete a simulation suite by ID.
Delete a subaccountTool to delete an existing subaccount.
Delete a teamTool to delete an existing team.
Detach actions from assistantTool to detach one or more actions from an AI assistant.
Detach knowledge baseTool to detach a knowledge base from an AI assistant.
Detach contact from memory storeTool to detach a contact from a memory store.
Detach memory store from agentTool to detach a memory store from an agent.
Execute simulation suiteTool to execute all test cases in a simulation suite.
Export analytics dataTool to export analytics data for calls within a specified date range.
Get action metadataTool to retrieve metadata about a specific action by its ID.
Get AI assistant detailsTool to retrieve details of a specific AI assistant.
Get phone call detailsTool to retrieve the transcript and detailed metadata for a specific phone call.
Get contact detailsTool to retrieve details of a specific contact by its ID.
Get knowledge baseTool to retrieve details of a specific knowledge base by its ID.
Get memory storeTool to retrieve details of a specific memory store by its ID.
Get memory store contact dataTool to retrieve memory data for a specific contact in a memory store.
Get phone numbersTool to retrieve a list of phone numbers associated with a workspace.
Get simulation detailsTool to retrieve details of a specific simulation by ID.
Get Simulation CaseTool to retrieve a simulation case by ID.
Get simulation scenarioTool to retrieve a simulation scenario by ID.
Get simulation suite by IDTool to retrieve a simulation suite by ID.
Get subaccount detailsTool to retrieve detailed metadata about a specific subaccount by ID.
Get team detailsTool to retrieve details of a specific team by its ID.
Initialize ActionTool to initialize a custom action with specified variables.
List actionsTool to list all actions in the workspace.
List AI assistantsTool to list all AI assistants associated with the account.
List call historyTool to retrieve call history (call logs) with filtering to check outcomes/statuses after placing calls.
List chatsTool to retrieve a list of chats, optionally filtered by agent ID.
List contactsTool to retrieve a list of contacts with optional search filtering.
List memory storesTool to list memory stores with optional filtering by title.
List Phone BooksTool to list all phone books in your workspace.
List Simulation CasesTool to list simulation cases with pagination and optional filtering by name or type.
List simulation cases by agentTool to list all simulation cases created for a specific agent.
List simulationsTool to list simulations with pagination and optional filters.
List simulation scenariosTool to list simulation scenarios with pagination and optional filtering.
List simulation sessionsTool to list simulation sessions with pagination and optional filters.
List simulation suitesTool to list simulation suites with pagination and optional filtering.
List subaccountsTool to list all subaccounts associated with the authenticated account.
List teamsTool to list assistant teams.
List voicesTool to list all text-to-speech voices in a workspace.
List webhook logsTool to retrieve paginated webhook logs with filtering and search capability.
Make a voice callTool to initiate a real-time voice call via the AI agent.
Start SimulationTool to start a new simulation using a simulation case.
Update ActionTool to update an existing action in Synthflow AI.
Update AssistantTool to update an existing assistant’s settings.
Update a contactTool to update an existing contact in Synthflow AI.
Update knowledge baseTool to update an existing knowledge base's name or usage conditions.
Update memory storeTool to update an existing memory store's title and description.
Update Simulation CaseTool to update an existing simulation case.
Update a simulation scenarioTool to update an existing simulation scenario.
Update an existing simulation suiteTool to update an existing simulation suite.
Update an existing teamTool to update an existing team.

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 you begin, make sure you have:
  • Node.js and npm installed
  • A Composio account with API key
  • An OpenAI API key

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 required dependencies

bash
npm install @ai-sdk/openai @ai-sdk/mcp @composio/core ai dotenv

First, install the necessary packages for your project.

What you're installing:

  • @ai-sdk/openai: Vercel AI SDK's OpenAI provider
  • @ai-sdk/mcp: MCP client for Vercel AI SDK
  • @composio/core: Composio SDK for tool integration
  • ai: Core Vercel AI SDK
  • dotenv: Environment variable management

Set up environment variables

bash
OPENAI_API_KEY=your_openai_api_key_here
COMPOSIO_API_KEY=your_composio_api_key_here
COMPOSIO_USER_ID=your_user_id_here

Create a .env file in your project root.

What's needed:

  • OPENAI_API_KEY: Your OpenAI API key for GPT model access
  • COMPOSIO_API_KEY: Your Composio API key for tool access
  • COMPOSIO_USER_ID: A unique identifier for the user session

Import required modules and validate environment

typescript
import "dotenv/config";
import { openai } from "@ai-sdk/openai";
import { Composio } from "@composio/core";
import * as readline from "readline";
import { streamText, type ModelMessage, stepCountIs } from "ai";
import { createMCPClient } from "@ai-sdk/mcp";

const composioAPIKey = process.env.COMPOSIO_API_KEY;
const composioUserID = process.env.COMPOSIO_USER_ID;

if (!process.env.OPENAI_API_KEY) throw new Error("OPENAI_API_KEY is not set");
if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set");
if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set");

const composio = new Composio({
  apiKey: composioAPIKey,
});
What's happening:
  • We're importing all necessary libraries including Vercel AI SDK's OpenAI provider and Composio
  • The dotenv/config import automatically loads environment variables
  • The MCP client import enables connection to Composio's tool server

Create Tool Router session and initialize MCP client

typescript
async function main() {
  // Create a tool router session for the user
  const session = await composio.create(composioUserID!, {
    toolkits: ["synthflow_ai"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Synthflow ai tools
  • The create method takes the user ID and specifies which toolkits should be available
  • The returned mcp object contains the URL and authentication headers needed to connect to the MCP server
  • This session provides access to all Synthflow ai-related tools through the MCP protocol

Connect to MCP server and retrieve tools

typescript
const mcpClient = await createMCPClient({
  transport: {
    type: "http",
    url: mcpUrl,
    headers: session.mcp.headers, // Authentication headers for the Composio MCP server
  },
});

const tools = await mcpClient.tools();
What's happening:
  • We're creating an MCP client that connects to our Composio Tool Router session via HTTP
  • The mcp.url provides the endpoint, and mcp.headers contains authentication credentials
  • The type: "http" is important - Composio requires HTTP transport
  • tools() retrieves all available Synthflow ai tools that the agent can use

Initialize conversation and CLI interface

typescript
let messages: ModelMessage[] = [];

console.log("Chat started! Type 'exit' or 'quit' to end the conversation.\n");
console.log(
  "Ask any questions related to synthflow_ai, like summarize my last 5 emails, send an email, etc... :)))\n",
);

const rl = readline.createInterface({
  input: process.stdin,
  output: process.stdout,
  prompt: "> ",
});

rl.prompt();
What's happening:
  • We initialize an empty messages array to maintain conversation history
  • A readline interface is created to accept user input from the command line
  • Instructions are displayed to guide the user on how to interact with the agent

Handle user input and stream responses with real-time tool feedback

typescript
rl.on("line", async (userInput: string) => {
  const trimmedInput = userInput.trim();

  if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) {
    console.log("\nGoodbye!");
    rl.close();
    process.exit(0);
  }

  if (!trimmedInput) {
    rl.prompt();
    return;
  }

  messages.push({ role: "user", content: trimmedInput });
  console.log("\nAgent is thinking...\n");

  try {
    const stream = streamText({
      model: openai("gpt-5"),
      messages,
      tools,
      toolChoice: "auto",
      stopWhen: stepCountIs(10),
      onStepFinish: (step) => {
        for (const toolCall of step.toolCalls) {
          console.log(`[Using tool: ${toolCall.toolName}]`);
          }
          if (step.toolCalls.length > 0) {
            console.log(""); // Add space after tool calls
          }
        },
      });

      for await (const chunk of stream.textStream) {
        process.stdout.write(chunk);
      }

      console.log("\n\n---\n");

      // Get final result for message history
      const response = await stream.response;
      if (response?.messages?.length) {
        messages.push(...response.messages);
      }
    } catch (error) {
      console.error("\nAn error occurred while talking to the agent:");
      console.error(error);
      console.log(
        "\nYou can try again or restart the app if it keeps happening.\n",
      );
    } finally {
      rl.prompt();
    }
  });

  rl.on("close", async () => {
    await mcpClient.close();
    console.log("\n👋 Session ended.");
    process.exit(0);
  });
}

main().catch((err) => {
  console.error("Fatal error:", err);
  process.exit(1);
});
What's happening:
  • We use streamText instead of generateText to stream responses in real-time
  • toolChoice: "auto" allows the model to decide when to use Synthflow ai tools
  • stopWhen: stepCountIs(10) allows up to 10 steps for complex multi-tool operations
  • onStepFinish callback displays which tools are being used in real-time
  • We iterate through the text stream to create a typewriter effect as the agent responds
  • The complete response is added to conversation history to maintain context
  • Errors are caught and displayed with helpful retry suggestions

Complete Code

Here's the complete code to get you started with Synthflow ai and Vercel AI SDK:

typescript
import "dotenv/config";
import { openai } from "@ai-sdk/openai";
import { Composio } from "@composio/core";
import * as readline from "readline";
import { streamText, type ModelMessage, stepCountIs } from "ai";
import { createMCPClient } from "@ai-sdk/mcp";

const composioAPIKey = process.env.COMPOSIO_API_KEY;
const composioUserID = process.env.COMPOSIO_USER_ID;

if (!process.env.OPENAI_API_KEY) throw new Error("OPENAI_API_KEY is not set");
if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set");
if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set");

const composio = new Composio({
  apiKey: composioAPIKey,
});

async function main() {
  // Create a tool router session for the user
  const session = await composio.create(composioUserID!, {
    toolkits: ["synthflow_ai"],
  });

  const mcpUrl = session.mcp.url;

  const mcpClient = await createMCPClient({
    transport: {
      type: "http",
      url: mcpUrl,
      headers: session.mcp.headers, // Authentication headers for the Composio MCP server
    },
  });

  const tools = await mcpClient.tools();

  let messages: ModelMessage[] = [];

  console.log("Chat started! Type 'exit' or 'quit' to end the conversation.\n");
  console.log(
    "Ask any questions related to synthflow_ai, like summarize my last 5 emails, send an email, etc... :)))\n",
  );

  const rl = readline.createInterface({
    input: process.stdin,
    output: process.stdout,
    prompt: "> ",
  });

  rl.prompt();

  rl.on("line", async (userInput: string) => {
    const trimmedInput = userInput.trim();

    if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) {
      console.log("\nGoodbye!");
      rl.close();
      process.exit(0);
    }

    if (!trimmedInput) {
      rl.prompt();
      return;
    }

    messages.push({ role: "user", content: trimmedInput });
    console.log("\nAgent is thinking...\n");

    try {
      const stream = streamText({
        model: openai("gpt-5"),
        messages,
        tools,
        toolChoice: "auto",
        stopWhen: stepCountIs(10),
        onStepFinish: (step) => {
          for (const toolCall of step.toolCalls) {
            console.log(`[Using tool: ${toolCall.toolName}]`);
          }
          if (step.toolCalls.length > 0) {
            console.log(""); // Add space after tool calls
          }
        },
      });

      for await (const chunk of stream.textStream) {
        process.stdout.write(chunk);
      }

      console.log("\n\n---\n");

      // Get final result for message history
      const response = await stream.response;
      if (response?.messages?.length) {
        messages.push(...response.messages);
      }
    } catch (error) {
      console.error("\nAn error occurred while talking to the agent:");
      console.error(error);
      console.log(
        "\nYou can try again or restart the app if it keeps happening.\n",
      );
    } finally {
      rl.prompt();
    }
  });

  rl.on("close", async () => {
    await mcpClient.close();
    console.log("\n👋 Session ended.");
    process.exit(0);
  });
}

main().catch((err) => {
  console.error("Fatal error:", err);
  process.exit(1);
});

Conclusion

You've successfully built a Synthflow ai agent using the Vercel AI SDK with streaming capabilities! This implementation provides a powerful foundation for building AI applications with natural language interfaces and real-time feedback.

Key features of this implementation:

  • Real-time streaming responses for a better user experience with typewriter effect
  • Live tool execution feedback showing which tools are being used as the agent works
  • Dynamic tool loading through Composio's Tool Router with secure authentication
  • Multi-step tool execution with configurable step limits (up to 10 steps)
  • Comprehensive error handling for robust agent execution
  • Conversation history maintenance for context-aware responses

You can extend this further by adding custom error handling, implementing specific business logic, or integrating additional Composio toolkits to create multi-app workflows.

How to build Synthflow ai MCP Agent with another framework

FAQ

What are the differences in Tool Router MCP and Synthflow ai MCP?

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

Can I use Tool Router MCP with Vercel AI SDK v6?

Yes, you can. Vercel AI SDK v6 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 Synthflow ai tools.

Can I manage the permissions and scopes for Synthflow ai while using Tool Router?

Yes, absolutely. You can configure which Synthflow ai 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 Synthflow ai 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.