How to integrate Calendly MCP with Vercel AI SDK

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

Introduction

This guide walks you through connecting Calendly to Vercel AI SDK using the Composio tool router. By the end, you'll have a working Calendly agent that can create a single-use scheduling link for my next meeting, cancel my 2pm event with a reason, mark an invitee as no-show for today's appointment, delete all invitee data for privacy compliance through natural language commands.

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

The Calendly MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Calendly account. It provides structured and secure access to your scheduling workflows, so your agent can perform actions like creating personalized scheduling links, managing events, handling invitee statuses, and automating reminders on your behalf.

  • Instant scheduling link creation: Direct your agent to generate single-use or shareable scheduling links so others can book time with you instantly—no more back-and-forth emails.
  • Automated event and invitee management: Have your agent cancel events, mark invitees as no-shows, or remove no-show statuses to keep your calendar accurate and up to date.
  • Custom one-off event setup: Empower your agent to create unique, one-off meeting types for special situations, bypassing your regular availability rules.
  • Webhook subscription automation: Let the agent set up webhook subscriptions to trigger notifications or workflows in real time when events happen in your Calendly account.
  • Data privacy and compliance actions: Instruct your agent to delete invitee data or scheduled event records as needed for privacy or regulatory compliance, especially for enterprise use cases.

Supported Tools & Triggers

Tools
Cancel eventPermanently cancels an existing, active scheduled event by its `uuid`, optionally providing a `reason`, which may trigger notifications to invitees.
Create an invitee no-showMarks an invitee, identified by their existing and valid uri, as a 'no show' for a scheduled event.
Create One-Off Event TypeCreates a temporary calendly one-off event type for unique meetings outside regular availability, requiring valid host/co-host uris, a future date/range for `date setting`, and a positive `duration`.
Create scheduling linkCreate a single-use scheduling link.
Create shareCreates a customizable, one-time share link for a calendly event type, allowing specific overrides to its settings (e.
Create single use scheduling linkCreates a one-time, single-use scheduling link for an active calendly event type, expiring after one booking.
Create webhook subscriptionCreates a calendly webhook subscription to notify a specified `url` (which must be a publicly accessible https endpoint) for selected `events` within a given `organization` and `scope`.
Delete invitee dataPermanently removes all invitee data associated with the provided emails from past organization events, for data privacy compliance (requires enterprise subscription; deletion may take up to one week).
Delete invitee no showDeletes an invitee no-show record by its `uuid` to reverse an invitee's 'no-show' status; the `uuid` must refer to an existing record.
Delete scheduled event dataFor enterprise users, initiates deletion of an organization's scheduled event data between a `start time` and `end time` (inclusive, where `start time` must be <= `end time`); actual data deletion may take up to 7 days to complete.
Delete webhook subscriptionDeletes an existing webhook subscription to stop calendly sending event notifications to its registered callback url; this operation is idempotent.
Get current userRetrieves detailed information about the currently authenticated calendly user.
Get eventUse to retrieve a specific calendly scheduled event by its uuid, provided the event exists in the user's calendly account.
Get event inviteeRetrieves detailed information about a specific invitee of a scheduled event, using their unique uuids.
Get event typeRetrieves details for a specific calendly event type, identified by its uuid, which must be valid and correspond to an existing event type.
Get groupRetrieves all attributes of a specific calendly group by its uuid; the group must exist.
Get group relationshipRetrieves a specific calendly group relationship by its valid and existing uuid, providing details on user-group associations and membership.
Get invitee no showRetrieves details for a specific invitee no show record by its uuid; an invitee no show is marked when an invitee does not attend a scheduled event.
Get organization invitationRetrieves a specific calendly organization invitation using its uuid and the parent organization's uuid.
Get organization membershipRetrieves a specific calendly organization membership by its uuid, returning all its attributes.
Get routing formRetrieves a specific routing form by its uuid, providing its configuration details including questions and routing logic.
Get userRetrieves comprehensive details for an existing calendly user.
Get user availability scheduleRetrieves an existing user availability schedule by its uuid; this schedule defines the user's default hours of availability.
Get webhook subscriptionRetrieves the details of an existing webhook subscription, identified by its uuid, including its callback url, subscribed events, scope, and state.
Invite user to organizationInvites a user to the specified calendly organization by email, if they aren't already a member and lack a pending invitation to it.
List activity log entriesRetrieves a list of activity log entries for a specified calendly organization (requires an active enterprise subscription), supporting filtering, sorting, and pagination.
List event inviteesRetrieves a list of invitees for a specified calendly event uuid, with options to filter by status or email, and sort by creation time.
List eventsRetrieves a list of scheduled calendly events; requires `user`, `organization`, `group`, or `invitee email` for scope, and admin rights may be needed when filtering by `organization` or `group`.
List event type available timesFetches available time slots for a calendly event type within a specified time range; results are not paginated.
List event type hostsRetrieves a list of hosts (users) assigned to a specific, existing calendly event type, identified by its uri.
List group relationshipsRetrieves a list of group relationships defining an owner's role (e.
List groupsReturns a list of groups for a specified calendly organization uri, supporting pagination.
List organization invitationsRetrieves a list of invitations for a specific organization, identified by its uuid.
List organization membershipsRetrieves a list of organization memberships.
List outgoing communicationsRetrieves a list of outgoing sms communications for a specified organization; requires an enterprise subscription and if filtering by creation date, both `min created at` and `max created at` must be provided to form a valid range.
List routing formsRetrieves routing forms for a specified organization; routing forms are questionnaires used to direct invitees to appropriate booking pages or external urls.
List user availability schedulesRetrieves all availability schedules for the specified calendly user.
List user busy timesFetches a user's busy time intervals (internal and external calendar events) in ascending order for a period up to 7 days; keyset pagination is not supported.
List user event typesRetrieves event types for a user or organization; requires either the `user` or `organization` uri.
List webhook subscriptionsRetrieves webhook subscriptions for a calendly organization; `scope` determines if `user` or `group` uri is also required for filtering.
Remove user from organizationRemoves a user (who is not an owner) from an organization by their membership uuid, requiring administrative privileges.
Revoke a user's organization invitationRevokes a pending and revokable (not yet accepted or expired) organization invitation using its uuid and the organization's uuid, rendering the invitation link invalid.

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: ["calendly"],
  });

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

  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 calendly, 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 Calendly 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 Calendly MCP Agent with another framework

OpenAI Agents SDK

OpenAI Agents SDK

Use Calendly MCP with OpenAI Agents SDK

Claude Agent SDK

Claude Agent SDK

Use Calendly MCP with Claude Agent SDK

Google ADK

Google ADK

Use Calendly MCP with Google ADK

LangChain

LangChain

Use Calendly MCP with LangChain

Mastra AI

Mastra AI

Use Calendly MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Calendly MCP with LlamaIndex

CrewAI

CrewAI

Use Calendly MCP with CrewAI

FAQ

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

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

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

Yes, absolutely. You can configure which Calendly 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 Calendly data and credentials are handled as safely as possible.

Used by agents from

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