How to integrate Postman MCP with Vercel AI SDK v6

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

Introduction

This guide walks you through connecting Postman to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Postman agent that can run the 'get users' request in postman, list all collections in my postman workspace, create a new environment variable in postman through natural language commands.

This guide will help you understand how to give your Vercel AI SDK agent real control over a Postman account through Composio's Postman 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 Postman integration
  • Using Composio's Tool Router to dynamically load and access Postman 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 Postman MCP server, and what's possible with it?

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

Supported Tools & Triggers

Tools
Create a CollectionTool to create a new Postman collection in a specific workspace or the default workspace.
Create a Collection CommentTool to create a comment on an API's collection.
Create Collection from SchemaTool to create a collection from a schema and link it to an API with specified relations.
Create a FolderTool to create a folder in a Postman collection.
Create a Folder CommentTool to create a comment on a folder.
Create a ForkTool to create a fork from an existing collection into a workspace.
Create Environment ForkTool to create a fork from an existing environment into a workspace.
Create a Mock ServerTool to create a new mock server in a Postman collection.
Create a MonitorTool to create a new monitor in a specific workspace to run a collection on a schedule.
Create an APITool to create a new API in Postman.
Create an APITool to create a new API in a Postman workspace.
Create an EnvironmentTool to create a new environment in a Postman workspace.
Create a Pull RequestTool to create a pull request for a forked collection into its parent collection.
Create Request in CollectionTool to create a new request in a Postman collection.
Create a Request CommentTool to create a comment on a request.
Create a ResponseTool to create a request response in a Postman collection.
Create a Response CommentTool to create a comment on a response.
Create API SchemaTool to create a schema for an API in Postman.
Create Mock Server ResponseTool to create a server response on a Postman mock server.
Create a SpecTool to create an API specification in Postman's Spec Hub.
Create Spec FileTool to create a new file in an API specification.
Create a WebhookTool to create a webhook that triggers a collection with a custom payload.
Create a WorkspaceTool to create a new workspace in Postman.
Create CollectionTool to create a new Postman collection with specified name and configuration.
Create EnvironmentTool to create a new Postman environment with specified name and variables.
Create Mock ServerTool to create a new mock server for a Postman collection.
Create MonitorTool to create a new monitor to run a collection on a schedule.
Create or Update a Schema FileTool to create or update an API schema file in Postman.
Create API Version RelationsTool to create new relations for an API version.
Create WebhookTool to create a new webhook for a Postman collection.
Delete a CollectionTool to permanently delete a collection from Postman.
Delete a collection's commentTool to delete a comment from an API's collection.
Delete a FolderTool to delete a folder in a Postman collection.
Delete a Folder's CommentTool to delete a comment from a folder.
Delete an APITool to delete an API from Postman.
Delete an API's CommentTool to delete a comment from an API.
Delete an environmentTool to delete an environment permanently in Postman.
Delete a Request's CommentTool to delete a comment from a request.
Delete a ResponseTool to delete a response in a Postman collection.
Delete a Response's CommentTool to delete a comment from a response.
Delete a Schema FileTool to delete a file in an API schema.
Delete Mock Server ResponseTool to delete a mock server's server response.
Delete a SpecTool to delete an API specification from Postman.
Delete Spec FileTool to delete a file from an API specification.
Delete a WorkspaceTool to delete a Postman workspace permanently.
Delete MonitorTool to delete a monitor by its ID.
Duplicate a CollectionTool to create a duplicate of a collection in another workspace.
Fork CollectionTool to create a fork of a collection in a specified workspace.
Generate Collection from SpecTool to generate a Postman collection from an OpenAPI 2.
Generate Spec from CollectionTool to generate an API specification from a Postman collection.
Get Billing Account DetailsTool to retrieve Postman billing account details for the authenticated team.
Get Collection CommentsTool to retrieve all comments left by users in an API's collection.
Get Collection's ForksTool to get all forks of a specific collection.
Get Collection Pull RequestsTool to get information about a collection's pull requests including source and destination IDs, status, and URLs.
Get Collection RolesTool to get information about all roles in a collection.
Get Folder InformationTool to retrieve information about a folder in a Postman collection.
Get Folder CommentsTool to retrieve all comments left by users in a folder.
Get All API ReleasesTool to get all releases for a specific API version in Postman.
Get All APIsTool to get all APIs accessible to the authenticated user with optional workspace filtering.
Get All APIs in WorkspaceTool to get all APIs in a specific Postman workspace.
Get All CollectionsTool to get all collections accessible to the authenticated user with optional workspace filtering.
Get All CollectionsTool to get all collections accessible to the authenticated user.
Get All EnvironmentsTool to get all environments accessible to the authenticated user with optional workspace filtering.
Get All Forked CollectionsTool to retrieve all forked collections for the authenticated user.
Get All GroupsTool to get all user groups in a Postman team.
Get All Linked RelationsTool to retrieve all linked relations for a specific API version in Postman.
Get All Mock ServersTool to get all active mock servers accessible to the authenticated user.
Get All MonitorsTool to get all monitors accessible to the authenticated user with optional workspace filtering.
Get All Workspace RolesTool to get information about all roles in a workspace based on the team's plan.
Get All API SpecificationsTool to get all API specifications in a workspace.
Get All Team UsersTool to get information about all users on the Postman team.
Get All Test RelationsTool to retrieve all test relations for a specific API version.
Get All API VersionsTool to get all published versions of a specific API in Postman.
Get All WorkspacesTool to get all workspaces accessible to the authenticated user with optional type filtering.
Get Monitor InformationTool to retrieve information about a specific monitor in Postman.
Get API InformationTool to retrieve information about a specific API in Postman.
Get API CommentsTool to retrieve all comments left by users in an API.
Get API VersionTool to get information about a specific API version in Postman.
Get an EnvironmentTool to retrieve detailed information about a specific environment in Postman.
Get Environment ForksTool to retrieve all forked environments for a specific environment.
Get Request InformationTool to retrieve information about a specific request in a Postman collection.
Get Request CommentsTool to retrieve all comments left by users in a request.
Get Response InformationTool to retrieve information about a saved response in a Postman collection.
Get Response CommentsTool to retrieve all comments left by users in a response.
Get API SchemaTool to retrieve information about an API schema from Postman.
Get API SpecificationTool to retrieve information about an API specification in Postman.
Get Spec File ContentsTool to get the contents of an API specification's file.
Get Spec DefinitionTool to get the complete contents of an API specification's definition.
Get Specification FilesTool to retrieve all files in an API specification from Postman.
Get Spec's Generated CollectionsTool to retrieve all collections generated from an API specification in Postman.
Get Async Collection Update StatusTool to get the status of an asynchronous collection update task.
Get Team UserTool to get information about a user on the Postman team.
Get Authenticated UserTool to get information about the authenticated user.
Get Workspace DetailsTool to get detailed information about a specific workspace by its ID.
Get Workspace Activity FeedTool to get a workspace's activity feed showing who added or removed collections, environments, or elements, and users joining or leaving.
Get Workspace RolesTool to get the roles of users, user groups, and partners in a workspace.
Get Collection Access KeysTool to retrieve all personal and team collection access keys for the authenticated user.
Get Contract Test RelationsTool to retrieve contract test relations for a specific API version.
Get documentation relationsTool to get documentation relations for a specific API version.
Get Duplication Task StatusTool to get the status of a collection duplication task.
Get Environment RelationsTool to get environment relations for a specific API version.
Get Generated SpecificationTool to retrieve the API specification generated for a Postman collection.
Get Workspace Global VariablesTool to get a workspace's global variables.
Get Integration Test RelationsTool to get integration test relations for a specific API version.
Get Resource TypesTool to get all resource types supported by Postman's SCIM API.
Get Schema File ContentsTool to get the contents of an API schema file at a specified path.
Get Schema FilesTool to retrieve files in an API schema from Postman.
Get Service Provider ConfigurationTool to get Postman's SCIM API service provider configuration information.
Get Source Collection StatusTool to check whether there is a change between a forked collection and its parent (source) collection.
Get Test Suite RelationsTool to get test suite relations for a specific API version.
Get Unclassified RelationsTool to get unclassified relations for an API version in Postman.
Import OpenAPI DefinitionTool to import an OpenAPI definition into Postman as a new collection.
Import OpenAPI SpecificationTool to import an OpenAPI specification into Postman as a new collection.
List Account InvoicesTool to get all invoices for a Postman billing account filtered by status.
Merge a ForkTool to merge a forked collection back into its parent collection.
Merge a forked environmentTool to merge a forked environment back into its parent environment.
Publish a Mock ServerTool to publish a mock server in Postman.
Pull Source Changes into ForkTool to pull changes from a parent (source) collection into a forked collection.
Replace an Environment's DataTool to completely replace an environment's data with new variables and values.
Replace Collection Data AsynchronouslyTool to replace the entire contents of a collection asynchronously.
Resolve a Comment ThreadTool to resolve a comment thread and any associated replies.
Review a Pull RequestTool to update the review status of a pull request by approving, declining, or unapproving it.
Run a MonitorTool to trigger an immediate run of a monitor and retrieve its execution results.
Sync Collection with API SchemaTool to sync a collection attached to an API with the API schema.
Sync Collection with SpecificationTool to sync a collection generated from an API specification.
Sync Spec with CollectionTool to sync an API specification with a linked collection.
Transfer FoldersTool to copy or move folders into a collection or folder.
Transform Collection to OpenAPITool to transform an existing Postman Collection into a stringified OpenAPI 3.
Update a FolderTool to update a folder in a Postman collection.
Update a Folder's CommentTool to update a comment on a folder.
Update a Mock ServerTool to update an existing mock server.
Update a MonitorTool to update an existing monitor in Postman.
Update an APITool to update an existing API in Postman.
Update an API's CommentTool to update a comment on an API.
Update an EnvironmentTool to update specific environment properties using JSON Patch operations (RFC 6902).
Update a Pull RequestTool to update an open pull request in Postman.
Update Request in CollectionTool to update a request in a Postman collection.
Update a Request's CommentTool to update a comment on a request.
Update a ResponseTool to update a response in a Postman collection.
Update a Response's CommentTool to update a comment on a response.
Update a Server ResponseTool to update a mock server's server response.
Update Spec FileTool to update an API specification file's content.
Update Spec PropertiesTool to update an API specification's properties such as its name.
Update a WorkspaceTool to update an existing workspace in Postman.
Update Workspace Global VariablesTool to update and replace a workspace's global variables.
Update Collection PropertiesTool to update specific collection properties like name, description, authentication, variables, or events.

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: ["postman"],
  });

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

  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 postman, 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("\nšŸ‘‹ Session ended.");
    process.exit(0);
  });
}

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

Conclusion

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

OpenAI Agents SDK

OpenAI Agents SDK

Use Postman MCP with OpenAI Agents SDK

Claude Agent SDK

Claude Agent SDK

Use Postman MCP with Claude Agent SDK

Google ADK

Google ADK

Use Postman MCP with Google ADK

LangChain

LangChain

Use Postman MCP with LangChain

Mastra AI

Mastra AI

Use Postman MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Postman MCP with LlamaIndex

CrewAI

CrewAI

Use Postman MCP with CrewAI

Claude Code

Claude Code

Use Postman MCP with Claude Code

Codex

Codex

Use Postman MCP with Codex

FAQ

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

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

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

Yes, absolutely. You can configure which Postman 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 Postman 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 callā†—Get started for free