How to integrate Kadoa MCP with Vercel AI SDK v6

Trusted by
AWS
Glean
Zoom
Airtable
GET STARTED FOR FREEGET A DEMO

30 min · no commitment · see it on your stack

Kadoa logo
Vercel AI SDK logo
divider

Introduction

This guide walks you through connecting Kadoa to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Kadoa agent that can fetch the latest data from your workflow, check crawl status for session abc123, list all pages crawled in last run through natural language commands.

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

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

Also integrate Kadoa with

TL;DR

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

The Kadoa MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Kadoa account. It provides structured and secure access to your data extraction workflows, so your agent can launch crawls, monitor sessions, retrieve extracted data, and manage workflow configurations automatically on your behalf.

  • Automated workflow monitoring and management: Ask your agent to fetch workflow configurations, enable data validation, or get the latest results from any extraction workflow you have set up.
  • Crawling session control: Have your agent check the status of crawl sessions, list all crawled pages, and pull the raw content (HTML or Markdown) from any page processed by a workflow.
  • Notification channel setup and retrieval: Direct your agent to create notification channels, list available notification event types, and fetch specific channel configurations for streamlined alerting.
  • Location and environment awareness: Let your agent retrieve all supported locations to ensure workflows run in the right environment before launching new extraction jobs.
  • Seamless data access: Instruct your agent to quickly get the most recent data output from any workflow, keeping your automations and dashboards always up to date.

Supported Tools & Triggers

Tools
Bulk Approve Validation RulesTool to bulk approve preview validation rules for a workflow.
Create Crawl ConfigTool to create a new crawling configuration in Kadoa.
Create Notification ChannelTool to create a notification channel for alerts delivery.
Create SchemaCreate a new data schema with specified fields and entity type.
Create Support IssueTool to create a support ticket in Kadoa.
Create Workflow TriggerTool to create a trigger that fires when a source workflow emits an event.
Delete All Validation RulesTool to soft-delete all validation rules for a specific workflow with optional audit trail.
Delete Crawl ConfigurationTool to delete a crawling configuration by its config ID.
Delete Notification ChannelTool to delete a notification channel by its ID.
Delete SchemaTool to delete a schema and all its revisions.
Delete Validation RuleTool to delete a validation rule from a Kadoa workflow.
Delete Validation Rules (Bulk)Tool to bulk delete multiple validation rules for a workflow.
Delete WorkflowDelete a workflow permanently from your Kadoa account.
Delete Workflow TriggerTool to delete a trigger from a Kadoa workflow.
Disable Validation RuleTool to disable a validation rule with a mandatory reason.
Enable Data ValidationTool to enable data validation on a specified workflow.
Execute Bulk Workflow OperationsExecute actions on multiple workflows at once.
Export Activity EventsTool to export activity events from audit logs to CSV format for compliance and audit purposes.
Export Activity WorkflowsTool to export workflow configurations and metadata as CSV for portfolio reviews and compliance reporting.
Get Workflow by IDRetrieve detailed configuration of a workflow by its ID.
Get all locationsRetrieves all available scraping proxy locations (countries) supported by Kadoa.
Get Crawl Bucket DataTool to retrieve file content from the Kadoa crawling bucket (HTML or screenshot).
Get Crawl ConfigurationTool to retrieve a crawling configuration by its ID.
Get Crawled Page ContentTool to retrieve content of a crawled page.
Get Crawled PagesTool to list pages crawled during a session.
Get Crawl StatusTool to fetch current status of a crawling session.
Get Event Type DetailsTool to retrieve details for a specific notification event type.
Get Notification Event TypesTool to retrieve supported notification event types.
Get Latest Workflow DataRetrieves the extracted data from a Kadoa workflow's most recent run (or a specific run if runId is provided).
Get Latest Workflow ValidationRetrieves the latest validation results for the most recent job of a workflow.
Get Notification ChannelTool to retrieve details of a specific notification channel.
Get Notification LogsTool to retrieve notification event logs with optional filtering by workflow, event type, and date range.
Get Notification SettingRetrieves a specific notification setting by its unique identifier.
Get Schema by IDRetrieve a specific schema by its unique identifier.
Get Validation AnomaliesTool to retrieve all anomalies for a specific validation.
Get Validation Anomalies By RuleTool to retrieve anomalies for a specific validation rule.
Get Validation ConfigurationTool to retrieve the data validation configuration for a specific workflow.
Get Validation RuleTool to retrieve a specific validation rule by its ID.
Get Workflow Audit LogRetrieve audit log entries for a workflow.
Get Workflow JobTool to retrieve the current status and telemetry information for a specific workflow job.
Get Workflow Run HistoryTool to fetch workflow run history.
Get WorkflowsRetrieve a paginated list of workflows with optional filtering.
Get Workflow TriggerTool to retrieve a specific trigger for a workflow.
Get Workflow Validation ResultsRetrieves the latest validation results for a specific workflow job.
Get Workspace DetailsTool to retrieve detailed information about a workspace (user, team, or organization).
List Activity EventsTool to retrieve activity events from audit logs with basic filtering and pagination.
List ChangesTool to retrieve all data changes detected across workflows in your Kadoa account.
List Crawl SessionsTool to retrieve a paginated list of crawling sessions with optional filtering.
List Job ValidationsTool to list all validation runs for a specific job with pagination support.
List Notification ChannelsTool to retrieve all notification channels configured for the account.
List Notification SettingsTool to retrieve all notification settings, with optional filtering by workflow ID or event type.
List SchemasTool to retrieve all schemas accessible by the authenticated user.
List Support StatesTool to retrieve available support issue states.
List Validation RulesTool to list all data validation rules with optional pagination and filtering.
List Workflow TriggersTool to get all triggers where the specified workflow is the source.
Pause Crawl SessionTool to pause an active crawling session.
Pause WorkflowTool to pause a running or scheduled workflow.
Create Advanced WorkflowTool to create an advanced workflow.
Start Crawl SessionStarts a new web crawling session to crawl and index pages from a website.
Create Notification SettingTool to create a notification setting linking channels to events.
Send Test NotificationSends a test notification event to verify notification channel configurations are working correctly.
Subscribe to Webhook EventsTool to subscribe to specified webhook events.
Create WorkflowCreate a new Kadoa web scraping workflow.
Configure Workflow MonitoringConfigure monitoring and scheduling for a Kadoa workflow to detect data changes.
Generate Workflow Validation RuleGenerate an AI-powered data validation rule for a Kadoa workflow.
Update Notification ChannelTool to update an existing notification channel.
Resume Crawl SessionTool to resume a paused crawling session.
Resume WorkflowResumes a paused, preview, or error workflow.
Run Ad-hoc ExtractionTool to synchronously extract data from a URL using a given template.
Run WorkflowTool to trigger a workflow to run immediately.
Schedule Validation JobTool to schedule a data validation job for a specific workflow job.
Unsubscribe from Webhook EventsUnsubscribe from webhook event notifications by deleting a notification setting.
Update Notification SettingsTool to update existing notification settings for events.
Update SchemaTool to update an existing Kadoa schema.
Update Validation ConfigurationTool to update the complete data validation configuration including alerting settings for a specific workflow.
Update Workflow MetadataTool to update workflow metadata such as name, description, tags, and configuration settings.
Update Workflow TriggerTool to update trigger properties including event type and enabled status.

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

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: ["kadoa"],
  });

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

  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 kadoa, 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 Kadoa 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 Kadoa MCP Agent with another framework

OpenAI Agents SDK logo

OpenAI Agents SDK

Use Kadoa MCP with OpenAI Agents SDK

Claude Agent SDK logo

Claude Agent SDK

Use Kadoa MCP with Claude Agent SDK

Claude Code logo

Claude Code

Use Kadoa MCP with Claude Code

Claude Cowork logo

Claude Cowork

Use Kadoa MCP with Claude Cowork

Codex logo

Codex

Use Kadoa MCP with Codex

OpenClaw logo

OpenClaw

Use Kadoa MCP with OpenClaw

Hermes logo

Hermes

Use Kadoa MCP with Hermes

CLI logo

CLI

Use Kadoa MCP with CLI

Google ADK logo

Google ADK

Use Kadoa MCP with Google ADK

LangChain logo

LangChain

Use Kadoa MCP with LangChain

Mastra AI logo

Mastra AI

Use Kadoa MCP with Mastra AI

LlamaIndex logo

LlamaIndex

Use Kadoa MCP with LlamaIndex

CrewAI logo

CrewAI

Use Kadoa MCP with CrewAI

FAQ

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

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

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

Yes, absolutely. You can configure which Kadoa 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 Kadoa 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.

Get started for freeGet a demo↗