How to integrate incident.io MCP with Vercel AI SDK v6

GET STARTED FOR FREE
Framework Integration Gradient
incident.io Logo
Vercel AI SDK Logo
divider

Introduction

This guide walks you through connecting incident.io to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working incident.io agent that can create a new incident for database outage, list all active incidents from this week, assign on-call responder to latest incident through natural language commands.

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

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

Supported Tools & Triggers

Tools
Create Alert Attribute V2Tool to create an alert attribute in incident.
Create Alert Route V2Tool to create an alert route in incident.
Create Alert Source V2Tool to create a new alert source in incident.
Create Catalog Entry V2Tool to create a catalog entry in incident.
Create Catalog Entry V3Tool to create a catalog entry in incident.
Create Catalog Type V3Tool to create a new catalog type in incident.
Create Custom Field OptionTool to create a new custom field option in incident.
Create Custom Field V2Tool to create a custom field in incident.
Create Escalation V2Tool to create an escalation in incident.
Create Incident RoleTool to create a new incident role in incident.
Create Incident Role V2Tool to create a new incident role in incident.
Create Incident StatusTool to create a new incident status in incident.
Create Incident V1Tool to create a new incident in incident.
Create Incident V2Tool to create a new incident in incident.
Create Managed Resource V2Tool to create a managed resource in incident.
Create SeverityTool to create a new severity level in incident.
Delete Alert Attribute V2Tool to delete an alert attribute from incident.
Delete Alert Route V2Tool to delete an alert route from incident.
Delete Alert Source V2Tool to delete an alert source from incident.
Delete Catalog Entry V2Tool to delete a catalog entry from incident.
Delete Catalog Entry V3Tool to archive a catalog entry from incident.
Delete Catalog Type V2Tool to delete a catalog type from incident.
Delete Catalog Type V3Tool to archive a catalog type and all its entries from incident.
Delete Custom FieldTool to delete a custom field from incident.
Delete Custom Field OptionTool to delete a custom field option in incident.
Delete Custom Field V2Tool to delete a custom field from incident.
Delete Escalation Path V2Tool to delete an escalation path from incident.
Delete Incident Role V2Tool to delete an incident role by ID.
Delete Incident Role V1Tool to delete an incident role by ID.
Delete Incident Status V1Tool to delete an incident status by its ID.
Delete Schedule V2Tool to delete a schedule from incident.
Delete SeverityTool to delete a severity in incident.
Delete Workflow V2Tool to delete a workflow from incident.
Edit Incident V2Tool to edit an existing incident in incident.
Show Alert Attribute V2Tool to retrieve a specific alert attribute by its ID from incident.
Show Alert Routes V2Tool to retrieve a specific alert route configuration by its ID from incident.
Show Alert Source V2Tool to retrieve a specific alert source by its ID from incident.
Show Entry Catalog V2Tool to retrieve a specific catalog entry by ID from incident.
Get Catalog Entry V3Tool to retrieve a specific catalog entry by ID from incident.
Get Catalog Type V2Tool to retrieve a specific catalog type by ID from incident.
Get Catalog Type V3Tool to retrieve a specific catalog type by ID from incident.
Get Custom Field OptionTool to retrieve a specific custom field option by its ID.
Get Custom Field V1Tool to retrieve details of a specific custom field by ID in incident.
Get Custom Field V2Tool to retrieve details of a specific custom field by ID using the V2 API in incident.
Show Escalations V2Tool to retrieve a specific escalation by ID from incident.
Get Incident RoleTool to retrieve a specific incident role by ID.
Get Incident Role V2Tool to retrieve a specific incident role by ID using V2 API.
Get Incident StatusTool to retrieve details of a specific incident status by ID.
Get Incident Timestamp V2Tool to retrieve a specific incident timestamp by ID using V2 API.
Get Incident TypeTool to retrieve detailed information about a specific incident type by ID.
Get Incident by ID (V1)Tool to retrieve a specific incident by ID.
Get Incident by ID (V2)Tool to retrieve a specific incident by ID using V2 API.
Show Schedules V2Tool to retrieve a specific schedule by ID from incident.
Get Severity V1Tool to retrieve a specific severity by ID from incident.
Get User V2Tool to retrieve a specific user by ID from incident.
Get Workflow V2Tool to retrieve a specific workflow by ID from incident.
List Actions V1Tool to list actions from incidents.
List Actions V2Tool to retrieve a list of actions from incident.
List Alert Attributes V2Tool to list all available alert attributes.
List Alert Routes V2Tool to list alert routes from incident.
List Alert Sources V2Tool to list all configured alert sources in incident.
List Alerts V2Tool to list alerts from incident.
List Catalog Entries V2Tool to list entries from a catalog type in incident.
List Catalog Entries V3Tool to list entries from a catalog type in incident.
List Catalog Resources V2Tool to retrieve all available catalog resource types.
List Catalog Resources V3Tool to retrieve all available catalog resource types from the V3 API.
List Catalog Types V2Tool to retrieve all catalog types configured for an organization.
List Catalog Types V3Tool to retrieve all catalog types configured for an organization using V3 API.
List Custom Field Options V1Tool to list all options for a specific custom field in incident.
List Custom Fields V1Tool to list all custom fields for an organization.
List Custom Fields V2Tool to list all custom fields V2 for an organization.
List Escalation Paths V2Tool to list escalation paths from incident.
List Escalations V2Tool to list escalations from incident.
List Follow-ups V2Tool to retrieve a list of follow-ups from incident.
Get API Key IdentityTool to retrieve identity information for the current API key.
List Incident Alerts V2Tool to list incident alerts from incident.
List Incident Attachments V1Tool to list incident attachments from incident.
List Incident Relationships V1Tool to list incident relationships for a given incident.
List Incident Roles V1Tool to list all incident roles for an organization.
List Incident Roles V2Tool to list all incident roles for an organization using V2 API.
List Incident Statuses V1Tool to list all incident statuses configured for an organization.
List Incidents V1Tool to list incidents from incident.
List Incidents V2Tool to list incidents from incident.
List Incident Timestamps V2Tool to list all incident timestamp configurations using V2 API.
List Incident Types V1Tool to list all incident types for an organization.
List Incident Updates V2Tool to list incident updates from incident.
List IP allowlistsTool to retrieve the current IP allowlist configuration for incident.
List Schedule Entries V2Tool to list schedule entries for a specific schedule in incident.
List Schedules V2Tool to list schedules from incident.
List Severities V1Tool to retrieve all incident severities configured for an organization.
List Status Pages V2Tool to list status pages from incident.
List Users V2Tool to list users in incident.
List Workflows V2Tool to list workflows from incident.
Update Alert Attribute V2Tool to update an alert attribute in incident.
Update Alert Source V2Tool to update an existing alert source configuration in incident.
Bulk Update Catalog Entries V3Tool to bulk update catalog entries in incident.
Update Catalog Entry V2Tool to update an existing catalog entry in incident.
Update Catalog Entry V3Tool to update an existing catalog entry in incident.
Update Catalog Type V2Tool to update a catalog type in incident.
Update Catalog Type V3Tool to update a catalog type in incident.
Update Custom Field Option V1Tool to update a custom field option in incident.
Update Custom Fields V1Tool to update an existing custom field configuration in incident.
Update Custom Field V2Tool to update an existing custom field in incident.
Update Escalation Path V2Tool to update an escalation path in incident.
Update Incident RoleTool to update an existing incident role in incident.
Update Incident Role V2Tool to update an existing incident role in incident.
Update Incident StatusTool to update an existing incident status in incident.
Update Schedule V2Tool to update an existing schedule in incident.
Update Catalog Type Schema V2Tool to update the schema of a catalog type in incident.
Update Catalog Type Schema V3Tool to update the schema of a catalog type in incident.
Update SeverityTool to update a severity in incident.
Update Workflow V2Tool to update an existing workflow in incident.

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

What is Tool Router?

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

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

How the Tool Router works

The Tool Router follows a three-phase workflow:

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

Step-by-step Guide

Prerequisites

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

Getting API Keys for OpenAI and Composio

OpenAI API Key
  • Go to the OpenAI dashboard and create an API key. You'll need credits to use the models, or you can connect to another model provider.
  • Keep the API key safe.
Composio API Key
  • Log in to the Composio dashboard.
  • Navigate to your API settings and generate a new API key.
  • Store this key securely as you'll need it for authentication.

Install required dependencies

bash
npm install @ai-sdk/openai @ai-sdk/mcp @composio/core ai dotenv

First, install the necessary packages for your project.

What you're installing:

  • @ai-sdk/openai: Vercel AI SDK's OpenAI provider
  • @ai-sdk/mcp: MCP client for Vercel AI SDK
  • @composio/core: Composio SDK for tool integration
  • ai: Core Vercel AI SDK
  • dotenv: Environment variable management

Set up environment variables

bash
OPENAI_API_KEY=your_openai_api_key_here
COMPOSIO_API_KEY=your_composio_api_key_here
COMPOSIO_USER_ID=your_user_id_here

Create a .env file in your project root.

What's needed:

  • OPENAI_API_KEY: Your OpenAI API key for GPT model access
  • COMPOSIO_API_KEY: Your Composio API key for tool access
  • COMPOSIO_USER_ID: A unique identifier for the user session

Import required modules and validate environment

typescript
import "dotenv/config";
import { openai } from "@ai-sdk/openai";
import { Composio } from "@composio/core";
import * as readline from "readline";
import { streamText, type ModelMessage, stepCountIs } from "ai";
import { createMCPClient } from "@ai-sdk/mcp";

const composioAPIKey = process.env.COMPOSIO_API_KEY;
const composioUserID = process.env.COMPOSIO_USER_ID;

if (!process.env.OPENAI_API_KEY) throw new Error("OPENAI_API_KEY is not set");
if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set");
if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set");

const composio = new Composio({
  apiKey: composioAPIKey,
});
What's happening:
  • We're importing all necessary libraries including Vercel AI SDK's OpenAI provider and Composio
  • The dotenv/config import automatically loads environment variables
  • The MCP client import enables connection to Composio's tool server

Create Tool Router session and initialize MCP client

typescript
async function main() {
  // Create a tool router session for the user
  const session = await composio.create(composioUserID!, {
    toolkits: ["incident_io"],
  });

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

  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 incident_io, 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 incident.io 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 incident.io MCP Agent with another framework

OpenAI Agents SDK

OpenAI Agents SDK

Use incident.io MCP with OpenAI Agents SDK

Claude Agent SDK

Claude Agent SDK

Use incident.io MCP with Claude Agent SDK

Google ADK

Google ADK

Use incident.io MCP with Google ADK

LangChain

LangChain

Use incident.io MCP with LangChain

Mastra AI

Mastra AI

Use incident.io MCP with Mastra AI

LlamaIndex

LlamaIndex

Use incident.io MCP with LlamaIndex

CrewAI

CrewAI

Use incident.io MCP with CrewAI

Claude Code

Claude Code

Use incident.io MCP with Claude Code

Codex

Codex

Use incident.io MCP with Codex

FAQ

What are the differences in Tool Router MCP and incident.io MCP?

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

Can I manage the permissions and scopes for incident.io while using Tool Router?

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

Used by agents from

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

Never worry about agent reliability

We handle tool reliability, observability, and security so you never have to second-guess an agent action.

Book a call↗Get started for free