How to integrate Shortcut MCP with Vercel AI SDK v6

This guide walks you through connecting Shortcut to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Shortcut agent that can create a new epic for q3 goals, add a threaded comment to our launch epic, start a new iteration for next sprint through natural language commands. This guide will help you understand how to give your Vercel AI SDK agent real control over a Shortcut account through Composio's Shortcut MCP server. Before we dive in, let's take a quick look at the key ideas and tools involved.

Shortcut logoShortcut
Api Key

Shortcut is a project management platform that aligns product development work with company objectives. It helps teams collaborate, track progress, and deliver on shared goals more efficiently.

136 Tools

Introduction

This guide walks you through connecting Shortcut to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Shortcut agent that can create a new epic for q3 goals, add a threaded comment to our launch epic, start a new iteration for next sprint through natural language commands.

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

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

Also integrate Shortcut with

TL;DR

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

The Shortcut MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Shortcut account. It provides structured and secure access to your project management workspace, so your agent can perform actions like creating epics, organizing teams, adding comments, managing sprints, and labeling tasks on your behalf.

  • Epic and story management: Direct your agent to create new epics, add threaded comments, or reply to existing discussions to streamline project tracking and collaboration.
  • Iteration and sprint planning: Let your agent set up new iterations, define sprint cycles, and associate tasks or groups, making agile planning and progress tracking a breeze.
  • Team and group organization: Effortlessly create and customize new groups or teams, assign members, and link workflows to keep your workspace structure organized and aligned.
  • Label and category creation: Ask your agent to generate new categories or labels, helping you organize stories, tasks, and priorities for efficient project management.
  • Linked file integration: Enable your agent to attach or create linked files within Shortcut, keeping all your project resources and documentation accessible and up to date.

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

What is Composio SDK?

Composio's Composio SDK helps agents find the right tools for a task at runtime. You can plug in multiple toolkits (like Gmail, HubSpot, and GitHub), and the agent will identify the relevant app and action to complete multi-step workflows. This can reduce token usage and improve the reliability of tool calls. Read more here: Getting started with Composio SDK

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

How the Composio SDK works

The Composio SDK follows a three-phase workflow:

  1. Discovery: Searches for tools matching your task and returns relevant toolkits with their details.
  2. Authentication: Checks for active connections. If missing, creates an auth config and returns a connection URL via Auth Link.
  3. Execution: Executes the action using the authenticated connection.

Step-by-step Guide

Step by step09 STEPS
1

Prerequisites

Before you begin, make sure you have:
  • Node.js and npm installed
  • A Composio account with API key
  • An OpenAI API key
2

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.
3

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
4

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
5

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
6

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: ["shortcut"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Shortcut 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 Shortcut-related tools through the MCP protocol
7

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 Shortcut tools that the agent can use
8

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 shortcut, 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
9

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 Shortcut 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 Shortcut 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: ["shortcut"],
  });

  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 shortcut, 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 Shortcut 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.
TOOLS

Supported Tools

Every Shortcut action and event your agent gets out of the box.

Create category

Create Category allows you to create a new Category in Shortcut.

Create doc

Create Doc allows you to create a new document in Shortcut.

Create entity template

Create a new entity template for the Workspace.

Create epic

Create Epic allows you to create a new Epic in Shortcut.

Create epic comment

This endpoint allows you to create a threaded Comment on an Epic.

Create epic comment comment

This endpoint allows you to create a nested Comment reply to an existing Epic Comment.

Create epic health

Create a new health status for the specified Epic.

Create generic integration

Tool to create a generic webhook integration in Shortcut.

Create group

Creates a new group in the Shortcut system with customizable properties.

Create iteration

The CreateIteration endpoint allows you to create a new Iteration in Shortcut, which represents a time-boxed period for completing a set of stories or tasks.

Create label

Create Label allows you to create a new Label in Shortcut.

Create linked file

Create Linked File allows you to create a new Linked File in Shortcut.

Create multiple stories

Create Multiple Stories allows you to create multiple stories in a single request using the same syntax as [Create Story](https://developer.

Create objective

Create Objective allows you to create a new Objective in Shortcut.

Create project

Create Project is used to create a new Shortcut Project.

Create story

Create Story is used to add a new story to your Shortcut Workspace.

Create story comment

Create Comment allows you to create a Comment on any Story.

Create story from template

Create Story From Template is used to add a new story derived from a template to your Shortcut Workspace.

Create story link

Story Links (called Story Relationships in the UI) allow you create semantic relationships between two stories.

Create story reaction

Create a reaction to a story comment.

Create task

Create Task is used to create a new task in a Story.

Delete category

Delete Category can be used to delete any Category.

Delete custom field

Deletes a specific custom field from the Shortcut system using its unique public identifier.

Delete doc

Permanently deletes a Doc and its associated data from Shortcut.

Delete entity template

Deletes a specific entity template from the Shortcut API.

Delete epic

Delete Epic can be used to delete the Epic.

Delete epic comment

This endpoint allows you to delete a Comment from an Epic.

Delete file

Delete File deletes a previously uploaded file.

Delete generic integration

Tool to delete a generic webhook integration from Shortcut.

Delete iteration

Deletes a specific iteration from the Shortcut project management system.

Delete label

Delete Label can be used to delete any Label.

Delete linked file

Delete Linked File can be used to delete any previously attached Linked-File.

Delete multiple stories

Delete Multiple Stories allows you to delete multiple archived stories at once.

Delete objective

Delete Objective can be used to delete any Objective.

Delete project

Delete Project can be used to delete a Project.

Delete story

Delete Story can be used to delete any Story.

Delete story comment

Delete a Comment from any story.

Delete story link

Removes the relationship between the stories for the given Story Link.

Delete story reaction

Delete a reaction from any story comment.

Delete task

Delete Task can be used to delete any previously created Task on a Story.

Disable iterations

Disables Iterations for the current workspace

Disable story templates

Disables the Story Template feature for the Workspace.

Enable iterations

Enables Iterations for the current workspace

Enable story templates

Enables the Story Template feature for the Workspace.

Get category

Get Category returns information about the selected Category.

Get current member info

Returns information about the authenticated member.

Get custom field

Retrieves detailed information about a specific custom field in Shortcut using its unique public identifier.

Get doc

Get Doc returns information about a chosen Doc by its public ID.

Get entity template

Get Entity Template returns information about a given entity template.

Get epic

Get Epic returns information about the selected Epic.

Get epic comment

This endpoint returns information about the selected Epic Comment.

Get epic health

Get the current health status for the specified Epic.

Get epic workflow

Returns the Epic Workflow for the Workspace.

Get external link stories

Get Stories which have a given External Link associated with them.

Get file

Get File returns information about the selected UploadedFile.

Get generic integration

Get Generic Integration returns information about a webhook integration by its ID.

Get group

Retrieves detailed information about a specific group in the Shortcut workspace using its unique public identifier.

Get iteration

Retrieves detailed information about a specific iteration in the Shortcut project management system.

Get key result

Get Key Result returns information about a chosen Key Result.

Get label

Get Label returns information about the selected Label.

Get linked file

Get File returns information about the selected Linked File.

Get member

Returns information about a Member.

Get objective

Get Objective returns information about a chosen Objective.

Get project

Get Project returns information about the selected Project.

Get repository

Get Repository returns information about the selected Repository.

Get story

Get Story returns information about a chosen Story.

Get story comment

Get Comment is used to get Comment information.

Get story link

Returns the stories and their relationship for the given Story Link.

Get task

Returns information about a chosen Task.

Get workflow

Get Workflow returns information about a chosen Workflow.

Link document to epic

Tool to create a relationship between a Document and an Epic.

List categories

List Categories returns a list of all Categories and their attributes.

List category milestones

List Category Milestones returns a list of all Milestones with the Category.

List category objectives

Returns a list of all Objectives with the Category.

List custom fields

Retrieves a list of all custom fields defined in the Shortcut API.

List docs

Tool to list Docs that the current user can read.

List document epics

Tool to retrieve all Epics related to a specific Document.

List entity templates

List all the entity templates for the Workspace.

List epic comments

Get a list of all Comments on an Epic.

List epic documents

Get a list of all Documents related to this Epic.

List epic healths

List the history of health statuses for the specified Epic, most recent first.

List epics

List Epics returns a list of all Epics and their attributes.

List epics paginated

List Epics with pagination returns a paginated list of Epics and their attributes.

List epic stories

Get a list of all Stories in an Epic.

List files

List Files returns a list of all UploadedFiles in the workspace.

List groups

A group in our API maps to a "Team" within the Shortcut Product.

List group stories

List the Stories assigned to the Group.

List iterations

Retrieves a list of all iterations in the Shortcut project management system.

List iteration stories

Get a list of all Stories in an Iteration.

List label epics

List all of the Epics with the Label.

List labels

List Labels returns a list of all Labels and their attributes.

List label stories

List all of the Stories with the Label.

List linked files

List Linked Files returns a list of all Linked-Files and their attributes.

List members

Returns information about members of the Workspace.

List members (slim)

Returns minimal member data optimized for Korey integration.

List objective epics

List all of the Epics within the Objective.

List objectives

List Objectives returns a list of all Objectives and their attributes.

List projects

List Projects returns a list of all Projects and their attributes.

List repositories

List Repositories returns a list of all Repositories and their attributes.

List stories

List Stories returns a list of all Stories in a selected Project and their attributes.

List story comment

Lists Comments associated with a Story

List story sub-tasks

Tool to list all Sub-task Stories for a given parent Story.

List workflows

Returns a list of all Workflows in the Workspace.

Search

Search lets you search Epics and Stories based on desired parameters.

Search documents

Tool to search Documents based on title and filters.

Search epics

Search Epics lets you search Epics based on desired parameters.

Search iterations

Search Iterations lets you search Iterations based on desired parameters.

Search milestones

Search Milestones lets you search Milestones based on desired parameters.

Search objectives

Search Objectives lets you search Objectives based on desired parameters.

Search stories

Search Stories lets you search Stories based on desired parameters.

Search stories old

Search Stories lets you search Stories based on desired parameters.

Story history

Retrieves the complete history of changes for a specific story in the Shortcut project management system.

Unlink comment thread from slack

Unlinks a Comment from its linked Slack thread (Comment replies and Slack replies will no longer be synced)

Unlink document from epic

Tool to remove the relationship between a Document and an Epic.

Unlink productboard from epic

This endpoint allows you to unlink a productboard epic.

Update category

Update Category allows you to replace a Category name with another name.

Update custom field

Update Custom Field can be used to update the definition of a Custom Field.

Update Doc

Tool to update an existing Doc's title and/or content.

Update entity template

Update an entity template's name or its contents.

Update epic

Update Epic can be used to update numerous fields in the Epic.

Update epic comment

This endpoint allows you to update a threaded Comment on an Epic.

Update file

Update File updates the properties of an UploadedFile (but not its content).

Update group

Updates an existing group in the Shortcut workspace.

Update health status

Update an existing health status by its ID.

Update iteration

This endpoint allows you to update an existing iteration in Shortcut.

Update key result

Update Key Result allows updating a Key Result's name or initial, observed, or target values.

Update label

Update Label allows you to replace a Label name with another name.

Update linked file

Updated Linked File allows you to update properties of a previously attached Linked-File.

Update multiple stories

Update Multiple Stories allows you to make changes to numerous stories at once.

Update objective

Update Objective can be used to update Objective properties.

Update project

Update Project can be used to change properties of a Project.

Update story

Update Story can be used to update Story properties.

Update story comment

Update Comment replaces the text of the existing Comment.

Update story link

Updates the stories and/or the relationship for the given Story Link.

Update task

Update Task can be used to update Task properties.

Upload files

Upload Files uploads one or many files and optionally associates them with a story.

FAQ

Frequently asked questions

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

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 Shortcut tools.

Yes, absolutely. You can configure which Shortcut 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.

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 Shortcut data and credentials are handled as safely as possible.

Start with Shortcut.It takes 30 seconds.

Managed auth, hosted MCP servers, and every Shortcut tool your agent needs.Free to start.

Start building