How to integrate Faraday MCP with Vercel AI SDK v6

GET STARTED FOR FREE
Framework Integration Gradient
Faraday Logo
Vercel AI SDK Logo
divider

Introduction

This guide walks you through connecting Faraday to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Faraday agent that can enrich salesforce leads with predictive insights, automate email categorization using faraday ai, trigger customer follow-up based on faraday scores through natural language commands.

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

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

TL;DR

Here's what you'll learn:
  • How to set up and configure a Vercel AI SDK agent with Faraday integration
  • Using Composio's Tool Router to dynamically load and access Faraday 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 Faraday MCP server, and what's possible with it?

The Faraday MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Faraday account. It provides structured and secure access so your agent can perform Faraday operations on your behalf.

Supported Tools & Triggers

Tools
Archive CohortTool to archive a cohort in Faraday.
Archive ConnectionTool to archive a connection in Faraday.
Archive DatasetTool to archive a dataset in Faraday.
Archive OutcomeTool to archive an outcome in Faraday.
Archive Persona SetTool to archive a persona set in Faraday.
Archive PlaceTool to archive a place in Faraday.
Archive ScopeTool to archive a scope in Faraday.
Archive StreamTool to archive a stream in Faraday.
Archive TargetTool to archive a target in Faraday.
Archive TraitTool to archive a trait in Faraday.
Create AccountTool to create a new Faraday account with specified name and branding.
Create CohortsTool to create a new cohort in Faraday.
Create DatasetTool to create a new dataset in Faraday.
Create OutcomeTool to create a new outcome.
Create Persona SetTool to create a new persona set.
Create PlaceTool to create a new place in Faraday.
Create ScopeTool to create a new scope in Faraday.
Create StreamTool to create a new stream or find an existing stream by name.
Create Target PreviewTool to start a preview delivery for a target.
Create TargetTool to create a target in Faraday.
Create TraitTool to create a new trait in Faraday.
Create Webhook EndpointTool to create a new webhook endpoint.
Delete AccountTool to delete an account by its UUID.
Delete All Orphaned TraitsTool to delete all orphaned traits from Faraday.
Delete CohortTool to permanently delete a cohort from Faraday.
Delete ConnectionTool to delete a Faraday connection by its UUID.
Delete DatasetTool to delete a dataset from Faraday.
Delete OutcomeTool to delete an outcome by its UUID.
Delete Persona SetTool to delete a persona set.
Delete PlaceTool to delete a place by its UUID.
Delete ScopeTool to delete a scope by its UUID.
Delete StreamTool to delete a stream by its UUID or name.
Delete TargetTool to delete a target by its UUID.
Delete UploadPermanently deletes a file from a Faraday dataset directory.
Delete Webhook EndpointTool to delete a webhook endpoint by its UUID.
Force Update CohortTool to trigger a rerun for a cohort.
Force Update ConnectionTool to trigger a rerun for a Faraday connection.
Force Update DatasetTool to trigger a rerun for a dataset.
Force Update OutcomeTool to trigger a rerun for an outcome resource.
Force Update Persona SetTool to trigger a rerun for a persona set resource.
Force Update PlaceTool to trigger a rerun for a place.
Force Update ScopeTool to trigger a rerun for a scope resource.
Force Update StreamTool to trigger a rerun for a stream.
Force Update TargetTool to trigger a rerun for a target.
Force Update TraitTool to trigger a rerun for a trait.
Get AccountTool to retrieve detailed information about a Faraday account by its UUID.
List AccountsTool to list all Faraday accounts accessible by the current user.
Get Accounts BillingTool to get billing information about your account.
Get CohortTool to retrieve a specific cohort by ID.
Get Cohort Membership AnalysisTool to get cohort membership counts over time for a specific cohort.
List CohortsTool to list all cohorts in your Faraday account.
Get Connection DatasetsTool to retrieve all datasets that use a specific connection.
Get Connection TargetsTool to retrieve all targets that use a specific connection.
Get Current AccountTool to retrieve the current account information.
Get Current Account BillingTool to retrieve billing information about the current account.
Get DatasetTool to retrieve a dataset by its UUID.
Get Dataset Ingress LogsTool to retrieve dataset ingress metrics over time.
List Feature StoresTool to list all feature stores.
Get Dependency GraphTool to retrieve the complete dependency graph for all resources on an account.
Get Market Opportunity AnalysesTool to retrieve all market opportunity analyses from Faraday.
Get OutcomeTool to retrieve an outcome by its ID.
List OutcomesTool to list all outcomes for the account.
Get Persona SetTool to retrieve a persona set by its UUID.
Get Persona Set Analysis FlowTool to retrieve the flow of persona sets over time, showing how individuals move between different personas.
List Persona SetsTool to list all persona sets.
Get Persona Set Analysis DimensionsTool to get various trait breakdown information about a persona set.
Get PlaceTool to retrieve a specific place by its UUID.
List PlacesTool to list all places in Faraday.
List RecommendersTool to list all recommenders.
Get ScopeTool to retrieve detailed information about a Faraday scope by its UUID.
Get Scope AnalysisTool to get analysis for a scope including outcomes and recommenders with probability distributions.
Get Scope DatasetsTool to retrieve all datasets associated with a Faraday scope.
Get Scope EfficacyTool to retrieve efficacy metrics for a scope.
Get Scope Payload CohortsTool to get payload cohorts for a specific scope.
Get Scope Payload OutcomesTool to retrieve payload outcomes for a specific scope.
Get Scope Payload Persona SetsTool to get payload persona sets for a specific scope.
Get Scope Payload RecommendersTool to retrieve payload recommenders for a specific scope.
Get Scope Population Exclusion CohortsTool to get population exclusion cohorts for a scope.
Get Scope Population CohortsTool to get population cohorts for a specific scope.
Get Scope TargetsTool to get all targets for a specific scope.
Get StreamTool to retrieve a stream by its UUID or name.
List StreamsTool to list all streams in your Faraday account.
Get Streams AnalysisTool to get the count of stream events emitted over a time period.
Get TargetTool to retrieve a specific target by its UUID from Faraday.
Get Target AnalysisTool to retrieve a target's analysis including geographic distributions and trait breakdowns.
Get TraitTool to retrieve a specific trait by ID.
Get Trait Analysis DimensionsTool to retrieve the percentage of the US population that falls into each category of a trait.
Get Traits CSVTool to retrieve all user-defined and Faraday-provided traits in CSV format.
Get UploadTool to download a previously uploaded file from a Faraday dataset directory.
List Uploaded FilesTool to retrieve the list of previously uploaded files in Faraday.
Get UsagesTool to retrieve usage statistics for your Faraday account.
Get Webhook EndpointTool to retrieve a webhook endpoint by its UUID.
List Webhook EndpointsTool to list all webhook endpoints configured for the account.
List AttributesTool to list all attributes in the feature store.
List ConnectionsTool to list all connections configured in Faraday.
List ScopesTool to list all scopes.
List TargetsTool to list all targets in your Faraday account.
Unarchive CohortTool to unarchive a previously archived cohort.
Unarchive DatasetTool to unarchive a dataset in Faraday.
Unarchive OutcomeTool to unarchive an outcome in Faraday.
Unarchive Persona SetTool to unarchive a previously archived persona set.
Unarchive PlaceTool to unarchive a previously archived place.
Unarchive ScopeTool to unarchive a previously archived scope.
Unarchive StreamTool to unarchive a stream in Faraday.
Unarchive TraitTool to unarchive a trait.
Update AccountTool to update an account's name or branding settings.
Update CohortTool to update a cohort's configuration using JSON Merge Patch semantics.
Update ConnectionTool to update a Faraday connection's name or options.
Update DatasetTool to update a dataset configuration using JSON Merge Patch semantics.
Update OutcomeTool to update an outcome's configuration using JSON Merge Patch semantics.
Update Persona SetTool to edit a persona set's configuration using JSON Merge Patch.
Update PlaceTool to update a place's name, addresses, or geojson geometry.
Update ScopeTool to update a Faraday scope's configuration using JSON Merge Patch semantics.
Update StreamTool to update a stream's properties using JSON Merge Patch semantics.
Update TargetTool to update a target's configuration in Faraday.
Update TraitTool to update a trait's properties using JSON Merge Patch semantics.
Update Webhook EndpointTool to update a webhook endpoint's configuration.

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

What is Tool Router?

Composio's Tool Router 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 Tool Router

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

How the Tool Router works

The Tool Router 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: ["faraday"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Faraday 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 Faraday-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 Faraday 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 faraday, 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("\nđź‘‹ 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 Faraday 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 Faraday 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: ["faraday"],
  });

  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 faraday, 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("\nđź‘‹ Session ended.");
    process.exit(0);
  });
}

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

Conclusion

You've successfully built a Faraday 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 Faraday MCP Agent with another framework

OpenAI Agents SDK

OpenAI Agents SDK

Use Faraday MCP with OpenAI Agents SDK

Claude Agent SDK

Claude Agent SDK

Use Faraday MCP with Claude Agent SDK

Google ADK

Google ADK

Use Faraday MCP with Google ADK

LangChain

LangChain

Use Faraday MCP with LangChain

Mastra AI

Mastra AI

Use Faraday MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Faraday MCP with LlamaIndex

CrewAI

CrewAI

Use Faraday MCP with CrewAI

Claude Code

Claude Code

Use Faraday MCP with Claude Code

Codex

Codex

Use Faraday MCP with Codex

FAQ

What are the differences in Tool Router MCP and Faraday MCP?

With a standalone Faraday MCP server, the agents and LLMs can only access a fixed set of Faraday tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Faraday 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 Faraday tools.

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

Yes, absolutely. You can configure which Faraday 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 Faraday 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.

Book a call↗Get started for free