How to integrate Salesforce MCP with Vercel AI SDK

BOOK A CALLGET STARTED FOR FREE
Framework Integration Gradient
Salesforce Logo
Vercel AI SDK Logo
divider

Introduction

This guide walks you through connecting Salesforce to Vercel AI SDK using the Composio tool router. By the end, you'll have a working Salesforce agent that can add new contact to spring campaign, clone opportunity with all associated products, complete follow-up task for lead smith, associate contact jane doe to acme account through natural language commands.

This guide will help you understand how to give your Vercel AI SDK agent real control over a Salesforce account through Composio's Salesforce 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 Salesforce integration
  • Using Composio's Tool Router to dynamically load and access Salesforce 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
  • Step Counting: Control multi-step tool execution
  • OpenAI Provider: Native integration with OpenAI models

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

The Salesforce MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Salesforce account. It provides structured and secure access to your CRM data, so your agent can perform actions like managing contacts, handling opportunities, automating campaigns, and tracking tasks on your behalf.

  • Automated contact and lead management: Effortlessly create new accounts, add contacts or leads to campaigns, and associate contacts with accounts to keep your CRM data up-to-date.
  • Streamlined opportunity management: Let your agent clone opportunities, add products to deals, and manage the full sales cycle for your pipeline.
  • Intelligent campaign automation: Enable your agent to create campaigns, enroll contacts or leads, and track campaign engagement for more effective marketing.
  • Task completion and workflow efficiency: Have your agent mark Salesforce tasks as completed and update records, keeping your team on track without manual intervention.
  • Flexible record operations: Allow the agent to clone existing records or apply lead assignment rules, ensuring data consistency and smart routing across your Salesforce environment.

Supported Tools & Triggers

Tools
Triggers
Add contact to campaignAdds a contact to a campaign by creating a campaignmember record, allowing you to track campaign engagement.
Add lead to campaignAdds a lead to a campaign by creating a campaignmember record, allowing you to track campaign engagement.
Add product to opportunityAdds a product (line item) to an opportunity.
Apply lead assignment rulesApplies configured lead assignment rules to a specific lead, automatically routing it to the appropriate owner based on your organization's rules.
Associate contact to accountAssociates a contact with an account by updating the contact's accountid field.
Clone opportunity with productsClones an opportunity and optionally its products (line items).
Clone recordCreates a copy of an existing salesforce record by reading its data, removing system fields, and creating a new record.
Complete taskMarks a task as completed with optional completion notes.
Create accountCreates a new account in salesforce with the specified information.
Create campaignCreates a new campaign in salesforce with the specified information.
Create contactCreates a new contact in salesforce with the specified information.
Create leadCreates a new lead in salesforce with the specified information.
Create noteCreates a new note attached to a salesforce record with the specified title and content.
Create opportunityCreates a new opportunity in salesforce with the specified information.
Create taskCreates a new task in salesforce to track activities, to-dos, and follow-ups related to contacts, leads, or other records.
Delete accountPermanently deletes an account from salesforce.
Delete campaignPermanently deletes a campaign from salesforce.
Delete contactPermanently deletes a contact from salesforce.
Delete leadPermanently deletes a lead from salesforce.
Delete notePermanently deletes a note from salesforce.
Delete opportunityPermanently deletes an opportunity from salesforce.
Get accountRetrieves a specific account by id from salesforce, returning all available fields.
Get campaignRetrieves a specific campaign by id from salesforce, returning all available fields.
Get contactRetrieves a specific contact by id from salesforce, returning all available fields.
Get dashboardGets detailed metadata for a specific dashboard including its components, layout, and filters.
Get leadRetrieves a specific lead by id from salesforce, returning all available fields.
Get noteRetrieves a specific note by id from salesforce, returning all available fields.
Get opportunityRetrieves a specific opportunity by id from salesforce, returning all available fields.
Get report metadataGets detailed metadata for a specific report including its structure, columns, filters, and groupings.
Get report instance resultsGets the results of a report instance created by running a report.
Get user infoRetrieves information about the current user or a specific user in salesforce.
List accountsLists accounts from salesforce using soql query, allowing flexible filtering, sorting, and field selection.
List campaignsLists campaigns from salesforce using soql query, allowing flexible filtering, sorting, and field selection.
List contactsLists contacts from salesforce using soql query, allowing flexible filtering, sorting, and field selection.
List dashboardsLists all dashboards available in salesforce with basic metadata including name, id, and urls.
List email templatesLists available email templates in salesforce with filtering and search capabilities.
List leadsLists leads from salesforce using soql query, allowing flexible filtering, sorting, and field selection.
List notesLists notes from salesforce using soql query, allowing flexible filtering, sorting, and field selection.
List opportunitiesLists opportunities from salesforce using soql query, allowing flexible filtering, sorting, and field selection.
List reportsLists all reports available in salesforce with basic metadata including name, id, and urls.
Log callLogs a completed phone call as a task in salesforce with call-specific details like duration, type, and disposition.
Log email activityCreates an emailmessage record to log email activity in salesforce, associating it with related records.
Mass transfer ownershipTransfers ownership of multiple records to a new owner in a single operation using salesforce's composite api for better performance.
Remove from campaignRemoves a lead or contact from a campaign by deleting the campaignmember record.
Run reportRuns a report and returns the results.
Run SOQL queryExecutes a soql query against salesforce data.
Search accountsSearch for salesforce accounts using multiple criteria like name, industry, type, location, or contact information.
Search campaignsSearch for salesforce campaigns using multiple criteria like name, type, status, date range, or active status.
Search contactsSearch for salesforce contacts using multiple criteria like name, email, phone, account, or title.
Search leadsSearch for salesforce leads using multiple criteria like name, email, phone, company, title, status, or lead source.
Search notesSearch for salesforce notes using multiple criteria like title, body content, parent record, owner, or creation date.
Search opportunitiesSearch for salesforce opportunities using multiple criteria like name, account, stage, amount, close date, or status.
Search tasksSearch for salesforce tasks using multiple criteria like subject, status, priority, assigned user, related records, or dates.
Send emailSends an email through salesforce with options for recipients, attachments, and activity logging.
Send email from templateSends an email using a predefined salesforce email template with merge field support.
Send mass emailSends bulk emails to multiple recipients, either using a template or custom content.
Update accountUpdates an existing account in salesforce with the specified changes.
Update campaignUpdates an existing campaign in salesforce with the specified changes.
Update contactUpdates an existing contact in salesforce with the specified changes.
Update leadUpdates an existing lead in salesforce with the specified changes.
Update noteUpdates an existing note in salesforce with the specified changes.
Update opportunityUpdates an existing opportunity in salesforce with the specified changes.
Update taskUpdates an existing task in salesforce with new information.

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 { experimental_createMCPClient as 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: ["salesforce"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Salesforce 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 Salesforce-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 Salesforce 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 salesforce, 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 Salesforce 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 Salesforce 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 { experimental_createMCPClient as 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: ["salesforce"],
  });

  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 salesforce, 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 Salesforce 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 Salesforce MCP Agent with another framework

OpenAI Agents SDK

OpenAI Agents SDK

Use Salesforce MCP with OpenAI Agents SDK

Claude Agent SDK

Claude Agent SDK

Use Salesforce MCP with Claude Agent SDK

Google ADK

Google ADK

Use Salesforce MCP with Google ADK

LangChain

LangChain

Use Salesforce MCP with LangChain

Mastra AI

Mastra AI

Use Salesforce MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Salesforce MCP with LlamaIndex

CrewAI

CrewAI

Use Salesforce MCP with CrewAI

FAQ

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

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

Yes, you can. Vercel AI SDK 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 Salesforce tools.

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

Yes, absolutely. You can configure which Salesforce 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 Salesforce 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