How to integrate Exa MCP with OpenAI Agents SDK

Connect OpenAI Agents SDK to Exa MCP. Summarize recent news articles on ai safety, find similar research papers to this url, and more using natural language, with authentication handled for you.

Exa logoExa
Api Key

Exa is a data extraction and search platform for gathering and analyzing information from websites, APIs, or databases. It helps teams quickly surface insights and automate data-driven workflows.

64 Tools

Introduction

This guide walks you through connecting Exa to the OpenAI Agents SDK using the Composio tool router. By the end, you'll have a working Exa agent that can search the web, generate answers, fetch page contents, build and enrich Websets, and set up search monitors through natural language commands.

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

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

Also integrate Exa with

TL;DR

Here's what you'll learn:
  • Get and set up your OpenAI and Composio API keys
  • Install the necessary dependencies
  • Initialize Composio and create a Tool Router session for Exa
  • Configure an AI agent that can use Exa as a tool
  • Run a live chat session where you can ask the agent to perform Exa operations

What is OpenAI Agents SDK?

The OpenAI Agents SDK is a lightweight framework for building AI agents that can use tools and maintain conversation state. It provides a simple interface for creating agents with hosted MCP tool support.

Key features include:

  • Hosted MCP Tools: Connect to external services through hosted MCP endpoints
  • SQLite Sessions: Persist conversation history across interactions
  • Simple API: Clean interface with Agent, Runner, and tool configuration
  • Streaming Support: Real-time response streaming for interactive applications

What is the Exa MCP server, and what's possible with it?

The Exa MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Exa account. It provides structured and secure access to Exa’s web search, content extraction, and research tools, so your agent can search the web, generate cited answers, retrieve website content, conduct in-depth research, and organize findings on your behalf.

  • Web search: Have your agent search the web with Exa to find relevant sources for a topic, question, company, or market.
  • Citation-backed answers: Ask the agent to answer specific questions or summarize broad topics using supporting web sources and citations.
  • Website content extraction: Direct your agent to retrieve text, highlights, summaries, and linked subpages from websites or saved documents.
  • Multi-step research: Let the agent start research tasks, check their progress, and return synthesized findings with citations when the work is complete.
  • Research collection management: Instruct your agent to create or delete web collections, import data into them, review imports, add enrichment, and set up monitors.

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 starting, make sure you have:
  • Composio API Key and OpenAI API Key
  • Primary know-how of OpenAI Agents SDK
  • A live Exa project
  • Some knowledge of Python or Typescript
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
3

Install dependencies

npm install @composio/openai-agents @openai/agents dotenv

Install the Composio SDK and the OpenAI Agents SDK.

4

Set up environment variables

bash
OPENAI_API_KEY=sk-...your-api-key
COMPOSIO_API_KEY=your-api-key
USER_ID=composio_user@gmail.com

Create a .env file and add your OpenAI and Composio API keys.

5

Import dependencies

import 'dotenv/config';
import { Composio } from '@composio/core';
import { OpenAIAgentsProvider } from '@composio/openai-agents';
import { Agent, hostedMcpTool, run, OpenAIConversationsSession } from '@openai/agents';
import * as readline from 'readline';
What's happening:
  • You're importing all necessary libraries.
  • The Composio and OpenAIAgentsProvider classes are imported to connect your OpenAI agent to Composio tools like Exa.
6

Set up the Composio instance

dotenv.config();

const composioApiKey = process.env.COMPOSIO_API_KEY;
const userId = process.env.USER_ID;

if (!composioApiKey) {
  throw new Error('COMPOSIO_API_KEY is not set. Create a .env file with COMPOSIO_API_KEY=your_key');
}
if (!userId) {
  throw new Error('USER_ID is not set');
}

// Initialize Composio
const composio = new Composio({
  apiKey: composioApiKey,
  provider: new OpenAIAgentsProvider(),
});
What's happening:
  • dotenv.config() loads your .env file so COMPOSIO_API_KEY and USER_ID are available as environment variables.
  • Creating a Composio instance using the API Key and OpenAIAgentsProvider class.
7

Create a Tool Router session

// Create Tool Router session for Exa
const session = await composio.create(userId as string, {
  toolkits: ['exa'],
});
const mcpUrl = session.mcp.url;

What is happening:

  • You give the Tool Router the user id and the toolkits you want available. Here, it is only exa.
  • The router checks the user's Exa connection and prepares the MCP endpoint.
  • The returned session.mcp.url is the MCP URL that your agent will use to access Exa.
  • This approach keeps things lightweight and lets the agent request Exa tools only when needed during the conversation.
8

Configure the agent

// Configure agent with MCP tool
const agent = new Agent({
  name: 'Assistant',
  model: 'gpt-5',
  instructions:
    'You are a helpful assistant that can access Exa. Help users perform Exa operations through natural language.',
  tools: [
    hostedMcpTool({
      serverLabel: 'tool_router',
      serverUrl: mcpUrl,
      headers: { 'x-api-key': composioApiKey },
      requireApproval: 'never',
    }),
  ],
});
What's happening:
  • We're creating an Agent instance with a name, model (gpt-5), and clear instructions about its purpose.
  • The agent's instructions tell it that it can access Exa and help with queries, inserts, updates, authentication, and fetching database information.
  • The tools array includes a hostedMcpTool that connects to the MCP server URL we created earlier.
  • The headers object includes the Composio API key for secure authentication with the MCP server.
  • requireApproval: 'never' means the agent can execute Exa operations without asking for permission each time, making interactions smoother.
9

Start chat loop and handle conversation

// Keep conversation state across turns
const conversationSession = new OpenAIConversationsSession();

// Simple CLI
const rl = readline.createInterface({
  input: process.stdin,
  output: process.stdout,
  prompt: 'You: ',
});

console.log('\nComposio Tool Router session created.');
console.log('\nChat started. Type your requests below.');
console.log("Commands: 'exit', 'quit', or 'q' to end\n");

try {
  const first = await run(agent, 'What can you help me with?', { session: conversationSession });
  console.log(`Assistant: ${first.finalOutput}\n`);
} catch (e) {
  console.error('Error:', e instanceof Error ? e.message : e, '\n');
}

rl.prompt();

rl.on('line', async (userInput) => {
  const text = userInput.trim();

  if (['exit', 'quit', 'q'].includes(text.toLowerCase())) {
    console.log('Goodbye!');
    rl.close();
    process.exit(0);
  }

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

  try {
    const result = await run(agent, text, { session: conversationSession });
    console.log(`\nAssistant: ${result.finalOutput}\n`);
  } catch (e) {
    console.error('Error:', e instanceof Error ? e.message : e, '\n');
  }

  rl.prompt();
});

rl.on('close', () => {
  console.log('\n👋 Session ended.');
  process.exit(0);
});
What's happening:
  • The program prints a session URL that you visit to authorize Exa.
  • After authorization, the chat begins.
  • Each message you type is processed by the agent using run().
  • The responses are printed to the console.
  • Typing exit, quit, or q cleanly ends the chat.

Complete Code

Here's the complete code to get you started with Exa and OpenAI Agents SDK:

import 'dotenv/config';
import { Composio } from '@composio/core';
import { OpenAIAgentsProvider } from '@composio/openai-agents';
import { Agent, hostedMcpTool, run, OpenAIConversationsSession } from '@openai/agents';
import * as readline from 'readline';

const composioApiKey = process.env.COMPOSIO_API_KEY;
const userId = process.env.USER_ID;

if (!composioApiKey) {
  throw new Error('COMPOSIO_API_KEY is not set. Create a .env file with COMPOSIO_API_KEY=your_key');
}
if (!userId) {
  throw new Error('USER_ID is not set');
}

// Initialize Composio
const composio = new Composio({
  apiKey: composioApiKey,
  provider: new OpenAIAgentsProvider(),
});

async function main() {
  // Create Tool Router session
  const session = await composio.create(userId as string, {
    toolkits: ['exa'],
  });
  const mcpUrl = session.mcp.url;

  // Configure agent with MCP tool
  const agent = new Agent({
    name: 'Assistant',
    model: 'gpt-5',
    instructions:
      'You are a helpful assistant that can access Exa. Help users perform Exa operations through natural language.',
    tools: [
      hostedMcpTool({
        serverLabel: 'tool_router',
        serverUrl: mcpUrl,
        headers: { 'x-api-key': composioApiKey },
        requireApproval: 'never',
      }),
    ],
  });

  // Keep conversation state across turns
  const conversationSession = new OpenAIConversationsSession();

  // Simple CLI
  const rl = readline.createInterface({
    input: process.stdin,
    output: process.stdout,
    prompt: 'You: ',
  });

  console.log('\nComposio Tool Router session created.');
  console.log('\nChat started. Type your requests below.');
  console.log("Commands: 'exit', 'quit', or 'q' to end\n");

  try {
    const first = await run(agent, 'What can you help me with?', { session: conversationSession });
    console.log(`Assistant: ${first.finalOutput}\n`);
  } catch (e) {
    console.error('Error:', e instanceof Error ? e.message : e, '\n');
  }

  rl.prompt();

  rl.on('line', async (userInput) => {
    const text = userInput.trim();

    if (['exit', 'quit', 'q'].includes(text.toLowerCase())) {
      console.log('Goodbye!');
      rl.close();
      process.exit(0);
    }

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

    try {
      const result = await run(agent, text, { session: conversationSession });
      console.log(`\nAssistant: ${result.finalOutput}\n`);
    } catch (e) {
      console.error('Error:', e instanceof Error ? e.message : e, '\n');
    }

    rl.prompt();
  });

  rl.on('close', () => {
    console.log('\nSession ended.');
    process.exit(0);
  });
}

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

Conclusion

This was a starter code for integrating Exa MCP with OpenAI Agents SDK to build a functional AI agent that can interact with Exa.

Key features:

  • Hosted MCP tool integration through Composio's Tool Router
  • SQLite session persistence for conversation history
  • Simple async chat loop for interactive testing
You can extend this by adding more toolkits, implementing custom business logic, or building a web interface around the agent.
TOOLS

Supported Tools

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

Generate an answer

Generates a direct, citation-backed answer to a clear natural language question or topic using Exa's search, adept at both specific answers and detailed summaries for open-ended queries.

Batch Search Monitors

Delete, pause, or unpause one batch of matching Search API monitors, not Websets monitors.

Cancel Agent Run

Cancel a queued or running Agent run immediately without returning results.

Cancel Webset

Cancel all operations on a Webset, stopping searches and enrichments.

Cancel Webset Enrichment

Cancel a running Webset enrichment and return its updated state.

Cancel Webset Search

Cancel a running search in a Webset and return its updated state.

Create Agent Run

Start an asynchronous Exa Agent run for multi-step web research or enrichment.

Create API Key

Create an API key for the authenticated Exa team.

Create Import

Tool to create a new import to upload data into a Webset.

Create a Monitor

Tool to create a new Monitor.

Create Search Monitor

Create a Search API monitor that delivers new, deduplicated search results to a webhook.

Create Webhook

Create a webhook delivering selected events to your URL.

Create Webset

Tool to create a new Webset with search, import, and enrichment setup.

Create Webset Enrichment

Create an enrichment task for items in an existing Webset.

Create Webset Search

Create a search for an existing Webset and return its initial state.

Delete Agent Run

Delete a stored Exa Agent run by ID and return its deletion receipt.

Delete API Key

Delete an API key belonging to the authenticated team using a Team Management service key.

Delete import

Tool to delete an existing import.

Delete Search Monitor

Permanently delete a Search API monitor and return the deleted monitor.

Delete Webhook

Delete a webhook, making its status inactive and stopping future deliveries to its URL.

Delete webset

Tool to delete a Webset.

Delete Webset Enrichment

Delete a Webset enrichment and return the deleted enrichment.

Delete Webset Item

Delete an item from a Webset and return the deleted item.

Delete Websets Monitor

Delete a Websets monitor and return the deleted monitor.

Get Agent Run

Retrieve an Agent run to inspect its status, output, citations, usage, or cost.

Get API Key

Tool to retrieve details of a specific API key by its ID.

Get API Key Usage

Get an Exa API key's total cost and cost breakdown for a time period.

Get contents from URLs or document IDs

Retrieve text / highlights / summary / subpages from URLs.

Get Event

Tool to get details of a specific event by its ID.

Get Import

Retrieve an import by ID to inspect its status, entity count, and failure details.

Get Search Monitor

Retrieve a Search API monitor's configuration, status, and next scheduled run by ID.

Get Search Monitor Run

Retrieve a Search API monitor run's status and full output when completed.

Get Webhook

Retrieve a webhook's status, events, destination URL, and metadata by ID.

Get Webset

Retrieve a Webset by ID or external ID, including its status and operations.

Get Webset Enrichment

Retrieve an enrichment configured on a Webset, including its status, description, format, and options.

Get Webset Item

Retrieve a Webset item by ID, including its properties, evaluations, and enrichments.

Get Webset Search

Retrieve a Webset search by ID to inspect its status, progress, and recall metrics.

Get Websets Monitor

Retrieve a Websets monitor's status, schedule, search behavior, and last run.

Get Websets Monitor Run

Retrieve a Websets monitor run's status, type, timestamps, and failure reason.

Get Websets Team Info

Get the authenticated Websets team's identity, current concurrency usage, and limits.

List Agent Run Events

List one page of stored Exa Agent run events as JSON.

List Agent Runs

List one page of your team's Exa Agent runs, ordered newest to oldest.

List API Keys

List the authenticated Exa team's API keys, or retrieve one by API key ID.

List events

Tool to list all events that have occurred in the system.

List imports

Tool to list all imports for the Webset.

List Search Monitor Runs

List one page of a Search API monitor's runs, newest first, including status and output.

List Search Monitors

List one page of the authenticated team's Search API monitors, not Websets monitors.

List Webhook Attempts

List one page of a webhook's delivery attempts in descending order.

List webhooks

Tool to list all webhooks for Websets.

List Webset Items

List one page of a Webset's items, optionally filtered by source ID.

List Websets

List one page of Websets, optionally filtered by ID, external ID, or title.

List Websets Monitor Runs

List one page of runs for a Websets monitor, including status and timestamps.

List Websets Monitors

List one page of Websets monitors, optionally filtered by Webset ID.

Preview Webset

Preview how Exa interprets a search before creating a Webset.

Search

Web search via the Exa engine.

Stop Agent Run

Complete a running ultra-effort Agent run early and return results gathered so far.

Trigger Search Monitor

Trigger an immediate Search API monitor run regardless of its schedule.

Update API Key

Update an API key's name, rate limit, or billing-period spending budget.

Update import

Tool to update an import configuration by ID.

Update Search Monitor

Update a Search API monitor's settings, pause or resume it, or remove its schedule with trigger=null.

Update Webhook

Update a webhook's destination URL, subscribed events, or metadata.

Update Webset

Update a Webset's title or metadata, leaving omitted fields unchanged.

Update Webset Enrichment

Update an enrichment's task, format, options, or metadata on a Webset.

Update Websets Monitor

Update a Websets monitor's status, schedule, search behavior, or metadata.

FAQ

Frequently asked questions

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

Yes, you can. OpenAI Agents 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 Exa tools.

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

Start with Exa.It takes 30 seconds.

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

Start building