How to integrate Celigo MCP with Autogen

This guide walks you through connecting Celigo to AutoGen using the Composio tool router. By the end, you'll have a working Celigo agent that can list failed celigo flows today, restart failed netsuite shopify syncs, check recent celigo job errors through natural language commands. This guide will help you understand how to give your AutoGen agent real control over a Celigo account through Composio's Celigo MCP server. Before we dive in, let's take a quick look at the key ideas and tools involved.

Get started for freeGet a demo
Celigo logoCeligo
Api Key

Celigo is an integration platform as a service for connecting apps, data, and business workflows. It helps teams automate cross-system processes without building every integration from scratch.

117 Tools

Introduction

This guide walks you through connecting Celigo to AutoGen using the Composio tool router. By the end, you'll have a working Celigo agent that can list failed celigo flows today, restart failed netsuite shopify syncs, check recent celigo job errors through natural language commands.

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

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

Also integrate Celigo with

OpenAI Agents SDKClaude Agent SDKClaude CodeClaude CoworkCodexOpenClawHermesCLIGoogle ADKLangChainVercel AI SDKMastra AILlamaIndexCrewAI

TL;DR

Here's what you'll learn:
  • Get and set up your OpenAI and Composio API keys
  • Install the required dependencies for Autogen and Composio
  • Initialize Composio and create a Tool Router session for Celigo
  • Wire that MCP URL into Autogen using McpWorkbench and StreamableHttpServerParams
  • Configure an Autogen AssistantAgent that can call Celigo tools
  • Run a live chat loop where you ask the agent to perform Celigo operations

What is AutoGen?

Autogen is a framework for building multi-agent conversational AI systems from Microsoft. It enables you to create agents that can collaborate, use tools, and maintain complex workflows.

Key features include:

  • Multi-Agent Systems: Build collaborative agent workflows
  • MCP Workbench: Native support for Model Context Protocol tools
  • Streaming HTTP: Connect to external services through streamable HTTP
  • AssistantAgent: Pre-built agent class for tool-using assistants

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

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

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 step08 STEPS
1

Prerequisites

You will need:

  • A Composio API key
  • An OpenAI API key (used by Autogen's OpenAIChatCompletionClient)
  • A Celigo account you can connect to Composio
  • Some basic familiarity with Autogen and Python async
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
  • 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.
3

Install dependencies

bash
pip install composio python-dotenv
pip install autogen-agentchat autogen-ext-openai autogen-ext-tools

Install Composio, Autogen extensions, and dotenv.

What's happening:

  • composio connects your agent to Celigo via MCP
  • autogen-agentchat provides the AssistantAgent class
  • autogen-ext-openai provides the OpenAI model client
  • autogen-ext-tools provides MCP workbench support

4

Set up environment variables

bash
COMPOSIO_API_KEY=your-composio-api-key
OPENAI_API_KEY=your-openai-api-key
USER_ID=your-user-identifier@example.com

Create a .env file in your project folder.

What's happening:

  • COMPOSIO_API_KEY is required to talk to Composio
  • OPENAI_API_KEY is used by Autogen's OpenAI client
  • USER_ID is how Composio identifies which user's Celigo connections to use
5

Import dependencies and create Tool Router session

python
import asyncio
import os
from dotenv import load_dotenv
from composio import Composio

from autogen_agentchat.agents import AssistantAgent
from autogen_ext.models.openai import OpenAIChatCompletionClient
from autogen_ext.tools.mcp import McpWorkbench, StreamableHttpServerParams

load_dotenv()

async def main():
    # Initialize Composio and create a Celigo session
    composio = Composio(api_key=os.getenv("COMPOSIO_API_KEY"))
    session = composio.create(
        user_id=os.getenv("USER_ID"),
        toolkits=["celigo"]
    )
    url = session.mcp.url
What's happening:
  • load_dotenv() reads your .env file
  • Composio(api_key=...) initializes the SDK
  • create(...) creates a Tool Router session that exposes Celigo tools
  • session.mcp.url is the MCP endpoint that Autogen will connect to
6

Configure MCP parameters for Autogen

python
# Configure MCP server parameters for Streamable HTTP
server_params = StreamableHttpServerParams(
    url=url,
    timeout=30.0,
    sse_read_timeout=300.0,
    terminate_on_close=True,
    headers={"x-api-key": os.getenv("COMPOSIO_API_KEY")}
)

Autogen expects parameters describing how to talk to the MCP server. That is what StreamableHttpServerParams is for.

What's happening:

  • url points to the Tool Router MCP endpoint from Composio
  • timeout is the HTTP timeout for requests
  • sse_read_timeout controls how long to wait when streaming responses
  • terminate_on_close=True cleans up the MCP server process when the workbench is closed
7

Create the model client and agent

python
# Create model client
model_client = OpenAIChatCompletionClient(
    model="gpt-5",
    api_key=os.getenv("OPENAI_API_KEY")
)

# Use McpWorkbench as context manager
async with McpWorkbench(server_params) as workbench:
    # Create Celigo assistant agent with MCP tools
    agent = AssistantAgent(
        name="celigo_assistant",
        description="An AI assistant that helps with Celigo operations.",
        model_client=model_client,
        workbench=workbench,
        model_client_stream=True,
        max_tool_iterations=10
    )

What's happening:

  • OpenAIChatCompletionClient wraps the OpenAI model for Autogen
  • McpWorkbench connects the agent to the MCP tools
  • AssistantAgent is configured with the Celigo tools from the workbench
8

Run the interactive chat loop

python
print("Chat started! Type 'exit' or 'quit' to end the conversation.\n")
print("Ask any Celigo related question or task to the agent.\n")

# Conversation loop
while True:
    user_input = input("You: ").strip()

    if user_input.lower() in ["exit", "quit", "bye"]:
        print("\nGoodbye!")
        break

    if not user_input:
        continue

    print("\nAgent is thinking...\n")

    # Run the agent with streaming
    try:
        response_text = ""
        async for message in agent.run_stream(task=user_input):
            if hasattr(message, "content") and message.content:
                response_text = message.content

        # Print the final response
        if response_text:
            print(f"Agent: {response_text}\n")
        else:
            print("Agent: I encountered an issue processing your request.\n")

    except Exception as e:
        print(f"Agent: Sorry, I encountered an error: {str(e)}\n")
What's happening:
  • The script prompts you in a loop with You:
  • Autogen passes your input to the model, which decides which Celigo tools to call via MCP
  • agent.run_stream(...) yields streaming messages as the agent thinks and calls tools
  • Typing exit, quit, or bye ends the loop

Complete Code

Here's the complete code to get you started with Celigo and AutoGen:

python
import asyncio
import os
from dotenv import load_dotenv
from composio import Composio

from autogen_agentchat.agents import AssistantAgent
from autogen_ext.models.openai import OpenAIChatCompletionClient
from autogen_ext.tools.mcp import McpWorkbench, StreamableHttpServerParams

load_dotenv()

async def main():
    # Initialize Composio and create a Celigo session
    composio = Composio(api_key=os.getenv("COMPOSIO_API_KEY"))
    session = composio.create(
        user_id=os.getenv("USER_ID"),
        toolkits=["celigo"]
    )
    url = session.mcp.url

    # Configure MCP server parameters for Streamable HTTP
    server_params = StreamableHttpServerParams(
        url=url,
        timeout=30.0,
        sse_read_timeout=300.0,
        terminate_on_close=True,
        headers={"x-api-key": os.getenv("COMPOSIO_API_KEY")}
    )

    # Create model client
    model_client = OpenAIChatCompletionClient(
        model="gpt-5",
        api_key=os.getenv("OPENAI_API_KEY")
    )

    # Use McpWorkbench as context manager
    async with McpWorkbench(server_params) as workbench:
        # Create Celigo assistant agent with MCP tools
        agent = AssistantAgent(
            name="celigo_assistant",
            description="An AI assistant that helps with Celigo operations.",
            model_client=model_client,
            workbench=workbench,
            model_client_stream=True,
            max_tool_iterations=10
        )

        print("Chat started! Type 'exit' or 'quit' to end the conversation.\n")
        print("Ask any Celigo related question or task to the agent.\n")

        # Conversation loop
        while True:
            user_input = input("You: ").strip()

            if user_input.lower() in ['exit', 'quit', 'bye']:
                print("\nGoodbye!")
                break

            if not user_input:
                continue

            print("\nAgent is thinking...\n")

            # Run the agent with streaming
            try:
                response_text = ""
                async for message in agent.run_stream(task=user_input):
                    if hasattr(message, 'content') and message.content:
                        response_text = message.content

                # Print the final response
                if response_text:
                    print(f"Agent: {response_text}\n")
                else:
                    print("Agent: I encountered an issue processing your request.\n")

            except Exception as e:
                print(f"Agent: Sorry, I encountered an error: {str(e)}\n")

if __name__ == "__main__":
    asyncio.run(main())

Conclusion

You now have an Autogen assistant wired into Celigo through Composio's Tool Router and MCP. From here you can:
  • Add more toolkits to the toolkits list, for example notion or hubspot
  • Refine the agent description to point it at specific workflows
  • Wrap this script behind a UI, Slack bot, or internal tool
Once the pattern is clear for Celigo, you can reuse the same structure for other MCP-enabled apps with minimal code changes.
TOOLS

Supported Tools

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

Assign Export Error

Tool to assign an export error to a user in Celigo.

Clone Export

Tool to clone a specific export to create a copy.

Clone Flow

Tool to clone a specific flow to create a copy.

Clone Import

Tool to clone a specific import to create a copy with remapped connections.

Clone Integration

Tool to clone a specific integration to create a copy.

Create Connection

Tool to create a new connection in Celigo to store credentials and application access information.

Create Export

Tool to create a new export in Celigo for extracting data from an application.

Create File Definition

Tool to create a file definition in Celigo for defining data file structure and format.

Create Flow

Tool to create a new flow in Celigo that composes export and import components together.

Create iClient

Tool to create a new iClient for SmartConnector authentication in Celigo.

Create Import

Tool to create a new import for inserting data into an application.

Create Integration

Tool to create a new integration in Celigo to group flows together.

Create Integration Revision

Tool to create a snapshot revision of a specific integration.

Create Script

Tool to create a new custom JavaScript script in Celigo for data transformation operations.

Create Tag

Tool to create a new tag in Celigo for organizing and categorizing resources.

Delete All Resource State

Tool to delete all state keys and values stored under a specific resource.

Delete All State

Tool to delete all global state keys and values in Celigo.

Delete Connection

Tool to delete a specific connection in Celigo.

Delete Export

Tool to delete a specific export from Celigo.

Delete Flow

Tool to delete a specific flow from Celigo.

Delete iClient

Tool to delete a specific iClient from Celigo.

Delete Import

Tool to delete a specific import from Celigo integrator.

Delete Integration

Tool to delete a specific integration from Celigo integrator.

Delete Resolved Errors

Tool to delete resolved errors for a specific export in Celigo.

Delete Resource State Value

Tool to delete the resource-specific key and value from state.

Delete Script

Tool to delete a specific script from Celigo.

Delete State Value

Tool to delete the specified state key and its value.

Delete Tag

Tool to delete a specific tag from Celigo integrator.

Delete User

Tool to delete a specific user from your Celigo account.

Evaluate Virtual Import Map

Evaluate the mapping step of a virtual import and return post-mapped records.

Execute Virtual Export

Execute a virtual export using a connection by posting the export model.

Execute Virtual Import

Tool to execute a virtual import by posting import configuration and data to a connection.

Virtual Export with Paging

Tool to execute a virtual export with paging support for Celigo connections.

Generate Structured File

Tool to generate sample EDI data or structured files from input data.

Get Connection by ID

Tool to get a specific connection by its ID.

Get Connection Audit Logs

Tool to retrieve audit logs for a specific connection.

Get Connection Debug Logs

Tool to retrieve debug logs for a connection by its ID.

Get Connection Dependencies

Tool to retrieve all resources using or used by a specific connection.

Get EDI Profiles

Tool to retrieve all EDI profiles for the account.

Get Export

Tool to retrieve a specific export by its ID from Celigo Integrator.

Get Export Audit Log

Tool to retrieve audit logs for a specific export in Celigo.

Get Export Errors

Tool to retrieve all open errors for a specific export within a flow.

Get File Definition

Tool to retrieve a specific file definition by its ID from Celigo Integrator.

Get File Definitions

Tool to retrieve all file definitions from Celigo.

Get Flow by ID

Tool to retrieve a specific Celigo flow by its ID.

Get Flow Audit Logs

Tool to retrieve audit logs for a specific flow.

Get Flow Dependencies

Tool to retrieve all resources using or used by a specific flow.

Get Flow Descendants

Tool to retrieve all descendant resources of a specific flow.

Get Flow Last Export Datetime

Tool to retrieve the last export datetime for a specific Celigo flow.

Get Flow Latest Jobs

Tool to retrieve the latest job executions for a specific flow.

Get Flow Template

Tool to download a Celigo flow as a template.

Get iClient by ID

Tool to get a specific iClient by its ID.

Get Import

Tool to retrieve a specific import by its ID from the Celigo Integrator API.

Get Import Audit Logs

Tool to retrieve audit logs for a specific import.

Get Import Dependencies

Tool to retrieve all resources using or used by a specific import.

Get Integration Audit Logs

Tool to retrieve audit logs for a specific integration.

Get Integration Connections

Tool to retrieve all connections associated with a specific integration.

Get Integration Dependencies

Tool to retrieve all resources using or used by a specific integration.

Get Integration Errors

Tool to retrieve all open errors for a specific integration in Celigo.

Get Integration Exports

Tool to retrieve all exports associated with a specific integration.

Get Integration Flows

Tool to retrieve all flows associated with a specific integration.

Get Integration Imports

Tool to retrieve all imports associated with a specific integration.

Get Integration Template

Tool to download a Celigo integration as a template.

Get Integration Users

Tool to retrieve users associated with a specific integration.

Get Job Errors

Tool to retrieve errors for a specific job in Celigo.

Get Resolved Errors

Tool to retrieve resolved errors for a specific export within a flow in Celigo.

Get Resource State Value

Tool to get the resource-specific value associated with a state key.

Get Script

Tool to retrieve a specific script by its ID from Celigo.

Get Script Logs

Tool to retrieve logs for a specific Celigo script by its ID.

Get State Value

Tool to get the value associated with a specified state key.

Get Tag

Tool to retrieve a specific tag by its ID from Celigo Integrator.

Get Tags

Tool to retrieve all tags from Celigo.

Get Token Info

Tool to retrieve information about the current API token.

Get User Share

Tool to retrieve a specific user share (ashare) by its ID.

Get All Users

Tool to retrieve all users from your Celigo account.

Invite Multiple Users

Tool to invite multiple users to a Celigo account with specified access levels.

Invite User

Tool to invite a user to a Celigo account with specified access levels.

Invoke Import

Tool to trigger/invoke a Celigo import by its ID.

List Audit Logs

Tool to retrieve all audit logs for resource changes.

List Connections

Tool to retrieve all connections in Celigo.

List Exports

Tool to retrieve all exports from Celigo integrator.

List Flows

Tool to retrieve all flows from Celigo.

List iClients

Tool to retrieve all iClients in Celigo.

List Imports

Tool to retrieve all imports from Celigo integrator.

List Integrations

Tool to retrieve all integrations from Celigo.

List Jobs

Fetches one page of jobs from Celigo with optional filters.

List Licenses

Tool to retrieve all SmartConnector licenses in Celigo.

List Resource State

Tool to retrieve all state keys for a specific resource in Celigo.

List State Keys

Tool to retrieve all global state keys from Celigo.

Parse CSV to JSON

Tool to convert CSV data to JSON format using Celigo's CSV parser.

Parse Structured File

Tool to convert EDI or structured file data (delimited or fixed-width) to JSON format.

Parse XML to JSON

Tool to convert XML data to JSON format using Celigo's XML parser processor.

Patch Connection

Tool to update part of a connection using JSON Patch operations (RFC 6902).

Patch Export

Tool to update specific fields of an export using JSON Patch operations.

Patch Flow

Tool to update part of a specific flow using JSON Patch operations (RFC 6902).

Patch iClient

Tool to update part of an iClient using JSON Patch operations (RFC 6902).

Patch Import

Tool to update specific fields of an import using JSON Patch operations.

Patch Integration

Tool to update part of an integration using JSON Patch operations (RFC 6902).

Ping Connection

Test that a specific Celigo connection is operational by pinging it.

Register Connection

Tool to register a connection to an integration in Celigo.

Register Connections Bulk

Tool to register multiple connections to an integration in Celigo at once.

Resolve Errors

Tool to resolve specified errors for an export in Celigo.

Retry Errors

Tool to retry specified errors for an export in Celigo.

Test Virtual Connection

Test a virtual connection where all details are submitted in the POST body but not saved.

Unregister Connection from Integration

Tool to unregister a connection from an integration in Celigo.

Update Connection

Tool to update a connection's configuration or credentials in Celigo.

Update Export

Tool to update an existing export configuration in Celigo.

Update Flow

Tool to update a specific flow's configuration in Celigo.

Update Flow Error Tags

Tool to update error tags for a specific flow export in Celigo.

Update iClient

Tool to update a specific iClient's configuration in Celigo.

Update Import

Tool to update a specific import's configuration in Celigo.

Update Integration

Tool to update a specific integration's configuration in Celigo.

Update Resource State Value

Tool to create or update the resource-specific state value for any Celigo resource type.

Update Script

Tool to update a specific script's code or configuration in Celigo.

Update State Value

Tool to create or update the value associated with a specified state key.

Update Tag

Tool to update a tag's name in Celigo.

Update User

Tool to update a specific user's account in Celigo.

FRAMEWORKS

How to build Celigo MCP Agent with another framework

OpenAI Agents SDK logo

OpenAI Agents SDK

Use Celigo MCP with OpenAI Agents SDK

Claude Agent SDK logo

Claude Agent SDK

Use Celigo MCP with Claude Agent SDK

Claude Code logo

Claude Code

Use Celigo MCP with Claude Code

Claude Cowork logo

Claude Cowork

Use Celigo MCP with Claude Cowork

Codex logo

Codex

Use Celigo MCP with Codex

OpenClaw logo

OpenClaw

Use Celigo MCP with OpenClaw

Hermes logo

Hermes

Use Celigo MCP with Hermes

CLI logo

CLI

Use Celigo MCP with CLI

Google ADK logo

Google ADK

Use Celigo MCP with Google ADK

LangChain logo

LangChain

Use Celigo MCP with LangChain

Vercel AI SDK logo

Vercel AI SDK

Use Celigo MCP with Vercel AI SDK

Mastra AI logo

Mastra AI

Use Celigo MCP with Mastra AI

LlamaIndex logo

LlamaIndex

Use Celigo MCP with LlamaIndex

CrewAI logo

CrewAI

Use Celigo MCP with CrewAI

FAQ

Frequently asked questions

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

Yes, you can. Autogen 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 Celigo tools.

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

Start with Celigo.It takes 30 seconds.

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

Start building
Customer stories

Teams ship real agents on Composio

GET STARTED FOR FREEBOOK A DEMO
agent.aiZoomLettaGleanHubSpotAWSWabi

Karan skipped his own birthday party to fix our critical issue. It was 10pm and he diverted his Waymo to help us instead. This really sets the bar — it shows the commitment you need when users rely on your software.

Harsha Gaddipati
Harsha Gaddipati
Co-founder, Slashy

A lot of students tell us that the moment their connected tools start talking to each other inside Opennote feels almost magical. The agent just knows them, and it's immensely helped keep new users on the platform.

Abhi Arya
Abhi Arya
Co-founder, Opennote

We chose Composio over Pipedream because it delivered depth where it mattered — niche tools and tricky edge cases other platforms simply ignored. That gave us confidence to scale without compromising.

Nirman Dave
Nirman Dave
CEO, Zams

As a solo builder, shipping fast is life or death. The only way I can outcompete incumbents is by outmanoeuvring them. Getting bogged down managing agent auth would have been a death sentence.

Ryan Yu
Ryan Yu
Founder, Extra Thursday

Before Composio, adding tool integrations was slow and resource-intensive. Each one could take weeks of engineering time, and maintaining them meant constantly keeping up with API changes.

Tomisin Jenrola
Tomisin Jenrola
Founder & CEO, SwarmZero

With hands-on help from their founder, we integrated Gmail and Google Drive in just 30 minutes. This level of personal support and commitment is exactly what startups should strive for.

Jerome Leclanche
Jerome Leclanche
Co-Founder, Ingram Technologies