How to integrate Attio MCP with LlamaIndex

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

30 min · no commitment · see it on your stack

Attio logo
LlamaIndex logo
divider

Introduction

This guide walks you through connecting Attio to LlamaIndex using the Composio tool router. By the end, you'll have a working Attio agent that can add a meeting note to john smith’s record, find all companies added this week, list recent notes for acme corp through natural language commands.

This guide will help you understand how to give your LlamaIndex agent real control over a Attio account through Composio's Attio MCP server.

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

Also integrate Attio with

TL;DR

Here's what you'll learn:
  • Set your OpenAI and Composio API keys
  • Install LlamaIndex and Composio packages
  • Create a Composio Tool Router session for Attio
  • Connect LlamaIndex to the Attio MCP server
  • Build a Attio-powered agent using LlamaIndex
  • Interact with Attio through natural language

What is LlamaIndex?

LlamaIndex is a data framework for building LLM applications. It provides tools for connecting LLMs to external data sources and services through agents and tools.

Key features include:

  • ReAct Agent: Reasoning and acting pattern for tool-using agents
  • MCP Tools: Native support for Model Context Protocol
  • Context Management: Maintain conversation context across interactions
  • Async Support: Built for async/await patterns

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

The Attio MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Attio account. It provides structured and secure access to your team’s relationship and workflow data, so your agent can perform actions like creating records, managing notes, searching your CRM, and organizing lists on your behalf.

  • Automated record management: Effortlessly create, update, or permanently delete records for people, companies, deals, and more in your Attio workspace.
  • Smart note taking and retrieval: Let your agent create, list, or delete notes attached to any record, keeping important context and meeting details organized for your team.
  • Powerful object and list navigation: Retrieve and explore all available objects or lists within your workspace, making it easy for your AI to understand and organize your CRM structure.
  • Advanced record search and filtering: Find specific CRM records either by unique ID or by searching with custom attributes, ensuring your agent surfaces the right data when you need it.
  • Comprehensive workspace insight: Get detailed information about object types and their attributes, so your AI-powered workflows always use the right fields and relationships.

Supported Tools & Triggers

Tools
Assert Company (Create or Update)Creates or updates a company record in Attio using a unique attribute to search for existing companies.
Assert Person RecordTool to create or update person records using a unique attribute to search for existing people.
Assert User Record (Create or Update)Creates or updates a user record in Attio using a unique attribute to search for existing users.
Assert Workspace (Create or Update)Creates or updates a workspace record in Attio using a unique attribute to search for existing workspaces.
Create AttributeTool to create a new attribute on an object or list in Attio.
Create CommentTool to create a new comment on a thread, record, or list entry in Attio.
Create CompanyCreates a new company record in Attio.
Create Deal RecordTool to create a new deal record in Attio.
Create ListTool to create a new list in Attio.
Create NoteThis tool creates a new note on a given record in Attio.
Create ObjectTool to create a new custom object in your Attio workspace.
Create PersonCreates a new person record in Attio.
Create RecordThis tool creates a new record in Attio for a specified object type (people, companies, deals, users, workspaces, etc.
Create Select OptionTool to add a new select option to a select or multiselect attribute in Attio.
Create StatusTool to add a new status to a status attribute on either an object or a list.
Create TaskTool to create a new task in Attio.
Create User RecordCreates a new user record in Attio.
Create WebhookTool to create a webhook and subscribe to events in Attio.
Create Workspace RecordCreates a new workspace record in Attio.
Delete CommentTool to delete a comment by its comment_id.
Delete CompanyTool to delete a company record from Attio by its record_id.
Delete DealTool to delete a deal record from Attio by its record_id.
Delete List EntryTool to delete a single list entry by its entry_id in Attio.
Delete NoteThis tool allows users to delete a specific note in Attio by its ID.
Delete PersonTool to delete a person record from Attio by its record_id.
Delete RecordThis tool allows you to delete a record from Attio permanently.
Delete TaskTool to delete a task by its task_id.
Delete UserTool to delete a user record from Attio by its record_id.
Delete WebhookTool to delete a webhook by its webhook_id.
Delete Workspace RecordTool to delete a workspace record from Attio by its record_id.
Find RecordThis tool allows users to find a record in Attio by either its unique ID or by searching using unique attributes.
Get Attribute DetailsTool to get information about a single attribute on either an object or a list.
Get CommentTool to get a single comment by its comment_id in Attio.
Get CompanyTool to get a single company record by its record_id in Attio.
Get Deal RecordTool to get a single deal record by its record_id in Attio.
Get ListTool to retrieve details of a single list in your Attio workspace.
Get List EntryTool to get a single list entry by its entry_id.
Get NoteTool to get a single note by its note_id in Attio.
Get ObjectTool to get a single object by its object_id or slug.
List Record EntriesTool to list all entries, across all lists, for which a record is the parent.
Get Current Token InfoTool to identify the current access token, the workspace it is linked to, and any permissions it has.
Get TaskTool to get a single task by its task_id in Attio.
List TasksTool to list all tasks in the workspace.
Get WebhookTool to get a single webhook by its webhook_id in Attio.
Get Workspace MemberTool to get a single workspace member by their workspace_member_id.
Get Workspace RecordTool to get a single workspace record by its record_id.
List Attribute OptionsTool to list all select options for a particular attribute on either an object or a list.
List AttributesTool to list the attribute schema for an Attio object or list (including slugs, types, select/status config) to enable correct filtering and writes.
List Attribute StatusesTool to list all statuses for a particular status attribute on either an object or a list.
List Call RecordingsTool to list all call recordings for a meeting in Attio.
List CompaniesTool to list company records in Attio with optional filtering and sorting.
List Company Attribute ValuesTool to get all values for a given attribute on a company record.
List Company Record EntriesTool to list all entries across all lists for which a company record is the parent in Attio.
List Deal EntriesTool to list all entries across all lists for which a deal record is the parent in Attio.
List Deal Record Attribute ValuesTool to retrieve all values for a specified attribute on a deal record in Attio.
List Deal RecordsTool to list deal records in Attio with the option to filter and sort results.
List List Entry Attribute ValuesTool to retrieve all values for a specified attribute on a list entry in Attio.
List ListsThis tool retrieves all lists available in the Attio workspace, sorted as they appear in the sidebar.
List MeetingsTool to list all meetings in the workspace using a deterministic sort order.
List NotesLists notes in Attio.
List ObjectsThis tool retrieves a list of all available objects (both system-defined and user-defined) in the Attio workspace via GET /v2/objects, returning key metadata including slugs and IDs for each object.
List People Attribute ValuesTool to get all values for a given attribute on a person record.
List People Record EntriesTool to list all entries across all lists for which a person record is the parent in Attio.
List Record Attribute ValuesTool to retrieve all values for a specified attribute on a record in Attio.
List RecordsThis tool lists records from a specific object type in Attio.
List ThreadsTool to list threads of comments on a record or list entry in Attio.
List User Record EntriesTool to list all entries across all lists for which a user record is the parent in Attio.
List User RecordsTool to list user records in Attio with optional filtering and sorting.
List WebhooksTool to get all webhooks in your Attio workspace.
List Workspace MembersTool to list workspace members (actors) so agents can reliably assign owners and resolve workspace-member IDs even when the optional Users standard object is disabled.
List Workspace Record Attribute ValuesTool to retrieve all values for a specified attribute on a workspace record in Attio.
List Workspace Record EntriesTool to list all entries across all lists for which a workspace record is the parent in Attio.
List Workspace RecordsTool to list workspace records with filtering and sorting options.
Patch RecordTool to update people, companies, and other records by record_id using PATCH method.
Update List Entry by Entry IDTool to update list entries by entry_id in Attio.
Get PersonTool to get a single person record by its record_id in Attio.
List Person RecordsTool to list person records from Attio with optional filtering and sorting.
Create List EntryTool to add a record to a list as a new list entry in Attio.
List List EntriesTool to list entries in a given list, with the option to filter and sort results.
List Records with QueryTool to list people, company or other records in Attio with the option to filter and sort results.
Assert List Entry by ParentTool to create or update a list entry for a given parent record in Attio.
Update List Entry by Entry ID (PUT)Tool to update list entries by entry_id in Attio using PUT method.
Assert Record (Create or Update)Tool to create or update people, companies and other records in Attio using a matching attribute.
Put Record (Overwrite)Tool to update people, companies, and other records by record_id using PUT method.
Query RecordsTool to query records for a specific Attio object using server-side filtering operators and sorting.
Search RecordsTool to fuzzy search for records across multiple objects in Attio.
Update AttributeTool to update an existing attribute by its attribute_id or slug.
Update CompanyTool to update a company record in Attio by its record_id.
Update Deal RecordTool to update an existing deal record in Attio by record ID.
Update ListTool to update an existing list in Attio.
Update ObjectTool to update a single object's configuration in Attio.
Update PersonTool to update a person record in Attio by its record_id.
Update RecordThis tool updates an existing record in Attio for a specified object type (people, companies, deals, users, workspaces, etc.
Update Select OptionTool to update an existing select option for a select or multiselect attribute in Attio.
Update StatusTool to update a status on a status attribute on either an object or a list in Attio.
Update TaskTool to update an existing task in Attio by its task_id.
Update User RecordTool to update a user record in Attio by its record_id.
Update WebhookTool to update a webhook's target URL and/or event subscriptions.
Update Workspace RecordTool to update a workspace record by ID using PATCH method.

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:
  • Python 3.8/Node 16 or higher installed
  • A Composio account with the API key
  • An OpenAI API key
  • A Attio account and project
  • Basic familiarity with async Python/Typescript

Getting API Keys for OpenAI, Composio, and Attio

OpenAI API key (OPENAI_API_KEY)
  • Go to the OpenAI dashboard
  • Create an API key if you don't have one
  • Assign it to OPENAI_API_KEY in .env
Composio API key and user ID
  • Log into the Composio dashboard
  • Copy your API key from Settings
    • Use this as COMPOSIO_API_KEY
  • Pick a stable user identifier (email or ID)
    • Use this as COMPOSIO_USER_ID

Installing dependencies

pip install composio-llamaindex llama-index llama-index-llms-openai llama-index-tools-mcp python-dotenv

Create a new Python project and install the necessary dependencies:

  • composio-llamaindex: Composio's LlamaIndex integration
  • llama-index: Core LlamaIndex framework
  • llama-index-llms-openai: OpenAI LLM integration
  • llama-index-tools-mcp: MCP client for LlamaIndex
  • python-dotenv: Environment variable management

Set environment variables

bash
OPENAI_API_KEY=your-openai-api-key
COMPOSIO_API_KEY=your-composio-api-key
COMPOSIO_USER_ID=your-user-id

Create a .env file in your project root:

These credentials will be used to:

  • Authenticate with OpenAI's GPT-5 model
  • Connect to Composio's Tool Router
  • Identify your Composio user session for Attio access

Import modules

import asyncio
import os
import dotenv

from composio import Composio
from composio_llamaindex import LlamaIndexProvider
from llama_index.core.agent.workflow import ReActAgent
from llama_index.core.workflow import Context
from llama_index.llms.openai import OpenAI
from llama_index.tools.mcp import BasicMCPClient, McpToolSpec

dotenv.load_dotenv()

Create a new file called attio_llamaindex_agent.py and import the required modules:

Key imports:

  • asyncio: For async/await support
  • Composio: Main client for Composio services
  • LlamaIndexProvider: Adapts Composio tools for LlamaIndex
  • ReActAgent: LlamaIndex's reasoning and action agent
  • BasicMCPClient: Connects to MCP endpoints
  • McpToolSpec: Converts MCP tools to LlamaIndex format

Load environment variables and initialize Composio

OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
COMPOSIO_API_KEY = os.getenv("COMPOSIO_API_KEY")
COMPOSIO_USER_ID = os.getenv("COMPOSIO_USER_ID")

if not OPENAI_API_KEY:
    raise ValueError("OPENAI_API_KEY is not set in the environment")
if not COMPOSIO_API_KEY:
    raise ValueError("COMPOSIO_API_KEY is not set in the environment")
if not COMPOSIO_USER_ID:
    raise ValueError("COMPOSIO_USER_ID is not set in the environment")

What's happening:

This ensures missing credentials cause early, clear errors before the agent attempts to initialise.

Create a Tool Router session and build the agent function

async def build_agent() -> ReActAgent:
    composio_client = Composio(
        api_key=COMPOSIO_API_KEY,
        provider=LlamaIndexProvider(),
    )

    session = composio_client.create(
        user_id=COMPOSIO_USER_ID,
        toolkits=["attio"],
    )

    mcp_url = session.mcp.url
    print(f"Composio MCP URL: {mcp_url}")

    mcp_client = BasicMCPClient(mcp_url, headers={"x-api-key": COMPOSIO_API_KEY})
    mcp_tool_spec = McpToolSpec(client=mcp_client)
    tools = await mcp_tool_spec.to_tool_list_async()

    llm = OpenAI(model="gpt-5")

    description = "An agent that uses Composio Tool Router MCP tools to perform Attio actions."
    system_prompt = """
    You are a helpful assistant connected to Composio Tool Router.
    Use the available tools to answer user queries and perform Attio actions.
    """
    return ReActAgent(tools=tools, llm=llm, description=description, system_prompt=system_prompt, verbose=True)

What's happening here:

  • We create a Composio client using your API key and configure it with the LlamaIndex provider
  • We then create a tool router MCP session for your user, specifying the toolkits we want to use (in this case, attio)
  • The session returns an MCP HTTP endpoint URL that acts as a gateway to all your configured tools
  • LlamaIndex will connect to this endpoint to dynamically discover and use the available Attio tools.
  • The MCP tools are mapped to LlamaIndex-compatible tools and plug them into the Agent.

Create an interactive chat loop

async def chat_loop(agent: ReActAgent) -> None:
    ctx = Context(agent)
    print("Type 'quit', 'exit', or Ctrl+C to stop.")

    while True:
        try:
            user_input = input("\nYou: ").strip()
        except (KeyboardInterrupt, EOFError):
            print("\nBye!")
            break

        if not user_input or user_input.lower() in {"quit", "exit"}:
            print("Bye!")
            break

        try:
            print("Agent: ", end="", flush=True)
            handler = agent.run(user_input, ctx=ctx)

            async for event in handler.stream_events():
                # Stream token-by-token from LLM responses
                if hasattr(event, "delta") and event.delta:
                    print(event.delta, end="", flush=True)
                # Show tool calls as they happen
                elif hasattr(event, "tool_name"):
                    print(f"\n[Using tool: {event.tool_name}]", flush=True)

            # Get final response
            response = await handler
            print()  # Newline after streaming
        except KeyboardInterrupt:
            print("\n[Interrupted]")
            continue
        except Exception as e:
            print(f"\nError: {e}")

What's happening here:

  • We're creating a direct terminal interface to chat with your Attio database
  • The LLM's responses are streamed to the CLI for faster interaction.
  • The agent uses context to maintain conversation history
  • You can type 'quit' or 'exit' to stop the chat loop gracefully
  • Agent responses and any errors are displayed in a clear, readable format

Define the main entry point

async def main() -> None:
    agent = await build_agent()
    await chat_loop(agent)

if __name__ == "__main__":
    # Handle Ctrl+C gracefully
    signal.signal(signal.SIGINT, lambda s, f: (print("\nBye!"), exit(0)))
    try:
        asyncio.run(main())
    except KeyboardInterrupt:
        print("\nBye!")

What's happening here:

  • We're orchestrating the entire application flow
  • The agent gets built with proper error handling
  • Then we kick off the interactive chat loop so you can start talking to Attio

Run the agent

npx ts-node llamaindex-agent.ts

When prompted, authenticate and authorise your agent with Attio, then start asking questions.

Complete Code

Here's the complete code to get you started with Attio and LlamaIndex:

import asyncio
import os
import signal
import dotenv

from composio import Composio
from composio_llamaindex import LlamaIndexProvider
from llama_index.core.agent.workflow import ReActAgent
from llama_index.core.workflow import Context
from llama_index.llms.openai import OpenAI
from llama_index.tools.mcp import BasicMCPClient, McpToolSpec

dotenv.load_dotenv()

OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
COMPOSIO_API_KEY = os.getenv("COMPOSIO_API_KEY")
COMPOSIO_USER_ID = os.getenv("COMPOSIO_USER_ID")

if not OPENAI_API_KEY:
    raise ValueError("OPENAI_API_KEY is not set")
if not COMPOSIO_API_KEY:
    raise ValueError("COMPOSIO_API_KEY is not set")
if not COMPOSIO_USER_ID:
    raise ValueError("COMPOSIO_USER_ID is not set")

async def build_agent() -> ReActAgent:
    composio_client = Composio(
        api_key=COMPOSIO_API_KEY,
        provider=LlamaIndexProvider(),
    )

    session = composio_client.create(
        user_id=COMPOSIO_USER_ID,
        toolkits=["attio"],
    )

    mcp_url = session.mcp.url
    print(f"Composio MCP URL: {mcp_url}")

    mcp_client = BasicMCPClient(mcp_url, headers={"x-api-key": COMPOSIO_API_KEY})
    mcp_tool_spec = McpToolSpec(client=mcp_client)
    tools = await mcp_tool_spec.to_tool_list_async()

    llm = OpenAI(model="gpt-5")
    description = "An agent that uses Composio Tool Router MCP tools to perform Attio actions."
    system_prompt = """
    You are a helpful assistant connected to Composio Tool Router.
    Use the available tools to answer user queries and perform Attio actions.
    """
    return ReActAgent(
        tools=tools,
        llm=llm,
        description=description,
        system_prompt=system_prompt,
        verbose=True,
    );

async def chat_loop(agent: ReActAgent) -> None:
    ctx = Context(agent)
    print("Type 'quit', 'exit', or Ctrl+C to stop.")

    while True:
        try:
            user_input = input("\nYou: ").strip()
        except (KeyboardInterrupt, EOFError):
            print("\nBye!")
            break

        if not user_input or user_input.lower() in {"quit", "exit"}:
            print("Bye!")
            break

        try:
            print("Agent: ", end="", flush=True)
            handler = agent.run(user_input, ctx=ctx)

            async for event in handler.stream_events():
                # Stream token-by-token from LLM responses
                if hasattr(event, "delta") and event.delta:
                    print(event.delta, end="", flush=True)
                # Show tool calls as they happen
                elif hasattr(event, "tool_name"):
                    print(f"\n[Using tool: {event.tool_name}]", flush=True)

            # Get final response
            response = await handler
            print()  # Newline after streaming
        except KeyboardInterrupt:
            print("\n[Interrupted]")
            continue
        except Exception as e:
            print(f"\nError: {e}")

async def main() -> None:
    agent = await build_agent()
    await chat_loop(agent)

if __name__ == "__main__":
    # Handle Ctrl+C gracefully
    signal.signal(signal.SIGINT, lambda s, f: (print("\nBye!"), exit(0)))
    try:
        asyncio.run(main())
    except KeyboardInterrupt:
        print("\nBye!")

Conclusion

You've successfully connected Attio to LlamaIndex through Composio's Tool Router MCP layer. Key takeaways:
  • Tool Router dynamically exposes Attio tools through an MCP endpoint
  • LlamaIndex's ReActAgent handles reasoning and orchestration; Composio handles integrations
  • The agent becomes more capable without increasing prompt size
  • Async Python provides clean, efficient execution of agent workflows
You can easily extend this to other toolkits like Gmail, Notion, Stripe, GitHub, and more by adding them to the toolkits parameter.

How to build Attio MCP Agent with another framework

ChatGPT logo

ChatGPT

Use Attio MCP with ChatGPT

Antigravity logo

Antigravity

Use Attio MCP with Antigravity

OpenAI Agents SDK logo

OpenAI Agents SDK

Use Attio MCP with OpenAI Agents SDK

Claude Agent SDK logo

Claude Agent SDK

Use Attio MCP with Claude Agent SDK

Claude Code logo

Claude Code

Use Attio MCP with Claude Code

Claude Cowork logo

Claude Cowork

Use Attio MCP with Claude Cowork

Codex logo

Codex

Use Attio MCP with Codex

Cursor logo

Cursor

Use Attio MCP with Cursor

OpenClaw logo

OpenClaw

Use Attio MCP with OpenClaw

Hermes logo

Hermes

Use Attio MCP with Hermes

CLI logo

CLI

Use Attio MCP with CLI

Google ADK logo

Google ADK

Use Attio MCP with Google ADK

LangChain logo

LangChain

Use Attio MCP with LangChain

Vercel AI SDK logo

Vercel AI SDK

Use Attio MCP with Vercel AI SDK

Mastra AI logo

Mastra AI

Use Attio MCP with Mastra AI

CrewAI logo

CrewAI

Use Attio MCP with CrewAI

OpenCode logo

OpenCode

Use Attio MCP with OpenCode

VS Code logo

VS Code

Use Attio MCP with VS Code

FAQ

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

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

Can I use Tool Router MCP with LlamaIndex?

Yes, you can. LlamaIndex 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 Attio tools.

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

Yes, absolutely. You can configure which Attio 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 Attio 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↗