How to integrate Asana MCP with Vercel AI SDK

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

Introduction

This guide walks you through connecting Asana to Vercel AI SDK using the Composio tool router. By the end, you'll have a working Asana agent that can create a new project for q3 goals, add followers to the product launch task, attach a file to today's meeting notes task, add a new color-coded tag to marketing workspace through natural language commands.

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

The Asana MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Asana account. It provides structured and secure access to your Asana workspace, so your agent can perform actions like creating tasks, managing projects, tagging work, assigning followers, and attaching files on your behalf.

  • Automated task creation and updates: Let your agent create new tasks with specific details, add tasks to projects or sections, and update them as work progresses.
  • Collaborator and follower management: Easily add users as followers to tasks, ensuring that the right people stay informed and engaged with project updates.
  • Project and workspace organization: Create new projects, define custom fields, and set up tags to keep your Asana workspace organized and tailored to your team's workflow.
  • File and attachment handling: Allow your agent to upload and attach important files directly to tasks, making sure all documentation stays in the right context.
  • Goal and resource linking: Link tasks, projects, or portfolios to goals, or add supporting relationships to help your team track progress and dependencies more effectively.

Supported Tools & Triggers

Tools
Triggers
Add Followers to TaskTool to add followers to a task in asana.
Add Supporting Relationship to GoalTool to add a supporting goal relationship to a goal.
Add task to sectionAdds an existing task to a section, optionally positioning it before or after another task in that section; if no position is specified, the task is added to the end.
Create AllocationCreates a new allocation.
Create a projectCreates a new asana project, requiring either a `workspace` or `team` gid for association, and returns the full project details.
Create a tag in a workspaceCreates a new tag, with properties like name and color defined in the request body, within a specific asana workspace (using `workspace gid`); this tag helps categorize tasks, is confined to the workspace, and is not automatically applied to tasks.
Create task in asana with specific detailsCreates a new asana task; requires 'workspace', 'parent', or 'projects' for association, and 'followers', 'projects', 'tags' are set only at creation.
Create Attachment for TaskTool to upload an attachment to a task.
Create Custom FieldTool to create a new custom field in a workspace.
Create Enum Option for Custom FieldTool to create a new enum option for a custom field in asana.
Create Project Status UpdateTool to create a new status update on a project.
Create a section in a projectCreates a new section in a project, optionally positioned relative to an existing section in the same project, and returns the full record of the new section.
Create subtaskCreates a new asana subtask under an existing parent task (`task gid`); `due on` and `due at` are mutually exclusive and cannot be set simultaneously.
Create task commentAdds a new text comment (story) to an existing asana task, appearing in its activity feed.
Create TeamTool to create a new team in an asana workspace.
Delete AllocationTool to delete an allocation by its id.
Delete AttachmentTool to delete an attachment by its globally unique identifier.
Delete Custom FieldTool to delete a custom field by its globally unique identifier.
Delete a projectDelete a project.
Delete a TagTool to delete a specific tag by its gid.
Delete a taskDelete a task.
Duplicate ProjectDuplicate a project.
Duplicate TaskDuplicate a task
Get AllocationGet an allocation by id.
Get AllocationsTool to get multiple allocations.
Get a projectRetrieves a specific asana project by its `project gid`, with an option to include additional fields for comprehensive details using `opt fields`; this action does not return tasks within the project.
Get a taskRetrieves full details for a specified task gid accessible by the user; use `opt fields` to customize returned data.
Get AttachmentTool to get a single attachment by its globally unique identifier.
Get Audit Log EventsTool to get audit log events for a workspace.
Get a user task listRetrieves a specific user's task list from asana by its `user task list gid`, optionally returning extended details like name, owner, and workspace if specified in `opt fields`.
Get current userRetrieves the authenticated user's full record, including accessible workspaces, often used as an initial call to establish user context for subsequent operations.
Get Custom FieldTool to get a single custom field by its globally unique identifier.
Get Custom Fields for WorkspaceTool to get all custom fields in a workspace.
Get Events on a ResourceRetrieve events on a resource to monitor changes.
Get GoalRetrieve the full record for a single goal by its gid.
Get Goal RelationshipsTool to retrieve goal relationships.
Get GoalsTool to retrieve multiple goals.
Get MembershipsTool to retrieve memberships for goals, projects, portfolios, or custom fields.
Get multiple projectsReturns a list of projects, optionally filtered by workspace, team, or archived status, supporting pagination for large datasets.
Get multiple tasksRetrieves a list of tasks, allowing filtering by assignee (requires `workspace`), project, section, `completed since`, and `modified since`; `workspace` also requires `assignee`.
Get multiple usersReturns a list of users in an asana workspace or organization, optionally filtered by workspace or team gid, with support for pagination and specifying optional fields.
Get multiple workspacesRetrieves all workspaces accessible by the authenticated user, returning an empty list if the user has no accessible workspaces.
Get PortfolioRetrieve the full record for a single portfolio by its gid.
Get Portfolio ItemsRetrieve items in a portfolio.
Get Portfolio MembershipsTool to retrieve multiple portfolio memberships.
Get PortfoliosRetrieve multiple portfolios.
Get Project BriefTool to retrieve a project's brief.
Get Project MembershipsTool to retrieve memberships from a project.
Get Projects for TeamTool to get a list of projects for a specific team in asana.
Get Project StatusTool to retrieve the full record for a single project status by its gid.
Get Project Status UpdatesTool to get status updates for a specific project.
Get Multiple Project TemplatesTool to retrieve multiple project templates.
Get SectionRetrieve the full record for a single section by its gid.
Get sections in a projectReturns compact records for all sections (used to group tasks) in a specified project.
Get Status UpdatesRetrieve status updates from an object.
Get Stories for TaskTool to get stories (comments, status updates, etc.
Get StoryTool to retrieve a story.
Get TagTool to get a single tag by its globally unique identifier.
Get TagsGet multiple tags in a workspace.
Get Task AttachmentsTool to get the list of attachments for a given task, project, or project brief.
Retrieve tasks for projectRetrieves tasks from a specified asana project, allowing filtering by completion status and selection of optional fields for detailed responses.
Get Task SubtasksTool to retrieve multiple task subtasks from a workspace.
Get Task TemplatesTool to retrieve multiple task templates from a workspace.
Get TeamTool to retrieve details of a specific team by its gid.
Get Team MembershipsTool to retrieve compact team membership records.
Get teams in workspaceReturns the compact records for all teams in the workspace visible to the authorized user.
Get Time PeriodsTool to retrieve compact or full representations of time periods.
Get Objects via TypeaheadTool to retrieve objects in a workspace via a typeahead search algorithm.
Get UserGet a user by their id.
Get Users for TeamGet users in a team.
Get Users in WorkspaceGet users in a workspace or organization.
Get WorkspaceTool to retrieve details of a specific workspace by its gid.
Get Workspace MembershipsTool to retrieve the workspace memberships for a specific workspace.
Get Workspace ProjectsTool to retrieve the projects associated with a specific workspace.
Insert or Reorder Enum Option for Custom FieldTool to reorder an existing enum option or insert a new enum option for a custom field at a specific position.
Remove Follower From TaskTool to remove one or more followers from a task.
Submit Parallel Requests (Batch API)Tool to submit multiple asana api requests in parallel using the batch api.
Update AllocationTool to update an existing allocation by its id.
Update a taskUpdates attributes of an existing asana task identified by its task gid.
Update Custom FieldTool to update a custom field by its globally unique identifier.
Update Enum OptionTool to update an enum option for a custom field.
Update a projectUpdate a project.
Update TagTool to update an existing tag by its globally unique identifier.
Update a teamTool to update details of an existing team.

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: ["asana"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Asana 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 Asana-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 Asana 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 asana, like summarize my last 5 emails, send an email, etc... :)))\n",
);

const rl = readline.createInterface({
  input: process.stdin,
  output: process.stdout,
  prompt: "> ",
});

rl.prompt();
What's happening:
  • We initialize an empty messages array to maintain conversation history
  • A readline interface is created to accept user input from the command line
  • Instructions are displayed to guide the user on how to interact with the agent

Handle user input and stream responses with real-time tool feedback

typescript
rl.on("line", async (userInput: string) => {
  const trimmedInput = userInput.trim();

  if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) {
    console.log("\nGoodbye!");
    rl.close();
    process.exit(0);
  }

  if (!trimmedInput) {
    rl.prompt();
    return;
  }

  messages.push({ role: "user", content: trimmedInput });
  console.log("\nAgent is thinking...\n");

  try {
    const stream = streamText({
      model: openai("gpt-5"),
      messages,
      tools,
      toolChoice: "auto",
      stopWhen: stepCountIs(10),
      onStepFinish: (step) => {
        for (const toolCall of step.toolCalls) {
          console.log(`[Using tool: ${toolCall.toolName}]`);
          }
          if (step.toolCalls.length > 0) {
            console.log(""); // Add space after tool calls
          }
        },
      });

      for await (const chunk of stream.textStream) {
        process.stdout.write(chunk);
      }

      console.log("\n\n---\n");

      // Get final result for message history
      const response = await stream.response;
      if (response?.messages?.length) {
        messages.push(...response.messages);
      }
    } catch (error) {
      console.error("\nAn error occurred while talking to the agent:");
      console.error(error);
      console.log(
        "\nYou can try again or restart the app if it keeps happening.\n",
      );
    } finally {
      rl.prompt();
    }
  });

  rl.on("close", async () => {
    await mcpClient.close();
    console.log("\nđź‘‹ Session ended.");
    process.exit(0);
  });
}

main().catch((err) => {
  console.error("Fatal error:", err);
  process.exit(1);
});
What's happening:
  • We use streamText instead of generateText to stream responses in real-time
  • toolChoice: "auto" allows the model to decide when to use Asana 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 Asana 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: ["asana"],
  });

  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 asana, like summarize my last 5 emails, send an email, etc... :)))\n",
  );

  const rl = readline.createInterface({
    input: process.stdin,
    output: process.stdout,
    prompt: "> ",
  });

  rl.prompt();

  rl.on("line", async (userInput: string) => {
    const trimmedInput = userInput.trim();

    if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) {
      console.log("\nGoodbye!");
      rl.close();
      process.exit(0);
    }

    if (!trimmedInput) {
      rl.prompt();
      return;
    }

    messages.push({ role: "user", content: trimmedInput });
    console.log("\nAgent is thinking...\n");

    try {
      const stream = streamText({
        model: openai("gpt-5"),
        messages,
        tools,
        toolChoice: "auto",
        stopWhen: stepCountIs(10),
        onStepFinish: (step) => {
          for (const toolCall of step.toolCalls) {
            console.log(`[Using tool: ${toolCall.toolName}]`);
          }
          if (step.toolCalls.length > 0) {
            console.log(""); // Add space after tool calls
          }
        },
      });

      for await (const chunk of stream.textStream) {
        process.stdout.write(chunk);
      }

      console.log("\n\n---\n");

      // Get final result for message history
      const response = await stream.response;
      if (response?.messages?.length) {
        messages.push(...response.messages);
      }
    } catch (error) {
      console.error("\nAn error occurred while talking to the agent:");
      console.error(error);
      console.log(
        "\nYou can try again or restart the app if it keeps happening.\n",
      );
    } finally {
      rl.prompt();
    }
  });

  rl.on("close", async () => {
    await mcpClient.close();
    console.log("\nđź‘‹ Session ended.");
    process.exit(0);
  });
}

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

Conclusion

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

OpenAI Agents SDK

OpenAI Agents SDK

Use Asana MCP with OpenAI Agents SDK

Claude Agent SDK

Claude Agent SDK

Use Asana MCP with Claude Agent SDK

Google ADK

Google ADK

Use Asana MCP with Google ADK

LangChain

LangChain

Use Asana MCP with LangChain

Mastra AI

Mastra AI

Use Asana MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Asana MCP with LlamaIndex

CrewAI

CrewAI

Use Asana MCP with CrewAI

FAQ

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

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

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

Yes, absolutely. You can configure which Asana 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 Asana 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