How to integrate Zep MCP with Autogen

GET STARTED FOR FREE
Framework Integration Gradient
Zep Logo
AutoGen Logo
divider

Introduction

This guide walks you through connecting Zep to AutoGen using the Composio tool router. By the end, you'll have a working Zep agent that can store a memory about today's meeting, retrieve all memories tagged urgent, summarize knowledge about client preferences through natural language commands.

This guide will help you understand how to give your AutoGen agent real control over a Zep account through Composio's Zep 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:
  • 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 Zep
  • Wire that MCP URL into Autogen using McpWorkbench and StreamableHttpServerParams
  • Configure an Autogen AssistantAgent that can call Zep tools
  • Run a live chat loop where you ask the agent to perform Zep 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 Zep MCP server, and what's possible with it?

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

Supported Tools & Triggers

Tools
Add Fact TripleTool to add a manually specified fact triple (subject-predicate-object) to the Zep knowledge graph.
Add Session MemoryTool to add memory messages to a specified Zep session.
Add Thread MessagesTool to add chat messages to a thread in Zep and ingest them into the user knowledge graph.
Clone GraphTool to clone a user or group graph with new identifiers in Zep.
Create GraphTool to create a new graph by adding data to Zep.
Create GroupTool to create a new group in Zep for multi-user graph management.
Create SessionTool to create a new session in Zep for storing conversation memory.
Create ThreadTool to create a new thread in Zep for a specific user.
Create UserTool to create a new user in Zep with properties like user_id, email, and metadata.
Delete GraphTool to delete a graph from Zep.
Delete GroupTool to delete a group from Zep.
Delete Session MemoryTool to delete a session and its memory from Zep.
Delete ThreadTool to delete a thread and its messages from Zep.
Delete UserTool to delete a user and all associated threads and artifacts from Zep.
Get Edge by UUIDTool to retrieve a specific edge by its UUID from the Zep knowledge graph.
Get Graph by IDTool to retrieve a graph by its unique identifier from Zep.
Get Group by IDTool to retrieve a group by ID from Zep.
Get Node Entity EdgesTool to retrieve all entity edges for a specific node in the Zep knowledge graph.
Get Project InfoTool to retrieve project information based on the provided API key.
Get Session by IDTool to retrieve a session by its unique identifier from Zep.
Get Session MemoryTool to retrieve memory for a given session including relevant facts and entities.
Get Session Message by UUIDTool to retrieve a specific message by UUID from a Zep session.
Get Session MessagesTool to retrieve messages for a given session from Zep.
Get Task StatusTool to check the status of asynchronous operations in Zep.
Get Thread MessagesTool to retrieve conversation history for a specific thread from Zep.
Get Thread User ContextTool to retrieve the most relevant user context from the user graph based on thread messages.
Get User by IDTool to retrieve a user by their user ID from Zep.
Get User NodeTool to retrieve a user's graph node and summary from Zep.
Get User NodesTool to retrieve all nodes for a specific user from their graph in Zep.
Get User SessionsTool to retrieve all sessions for a user from Zep.
Get User ThreadsTool to retrieve all threads for a specific user from Zep.
Graph SearchTool to perform hybrid graph search combining semantic similarity and BM25 full-text search across the Zep knowledge graph.
List GraphsTool to retrieve all graphs from Zep with pagination support.
List Groups OrderedTool to retrieve all groups from Zep with pagination support.
List Sessions OrderedTool to retrieve all sessions from Zep with pagination and ordering support.
List ThreadsTool to retrieve all threads from Zep with pagination support.
List Users OrderedTool to retrieve all users from Zep with pagination support.
List All ThreadsTool to list all threads with pagination and ordering support.
Update GraphTool to update graph information in Zep including name and description.
Update GroupTool to update group information in Zep including name, description, and fact rating instructions.
Update MessageTool to update a message in a Zep thread.
Update Session MetadataTool to update session metadata in Zep.
Update UserTool to update an existing user's information in Zep including email, metadata, and ontology settings.

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

You will need:

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

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 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 Zep via MCP
  • autogen-agentchat provides the AssistantAgent class
  • autogen-ext-openai provides the OpenAI model client
  • autogen-ext-tools provides MCP workbench support

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 Zep connections to use

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 Zep session
    composio = Composio(api_key=os.getenv("COMPOSIO_API_KEY"))
    session = composio.create(
        user_id=os.getenv("USER_ID"),
        toolkits=["zep"]
    )
    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 Zep tools
  • session.mcp.url is the MCP endpoint that Autogen will connect to

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

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 Zep assistant agent with MCP tools
    agent = AssistantAgent(
        name="zep_assistant",
        description="An AI assistant that helps with Zep 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 Zep tools from the workbench

Run the interactive chat loop

python
print("Chat started! Type 'exit' or 'quit' to end the conversation.\n")
print("Ask any Zep 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 Zep 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 Zep 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 Zep session
    composio = Composio(api_key=os.getenv("COMPOSIO_API_KEY"))
    session = composio.create(
        user_id=os.getenv("USER_ID"),
        toolkits=["zep"]
    )
    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 Zep assistant agent with MCP tools
        agent = AssistantAgent(
            name="zep_assistant",
            description="An AI assistant that helps with Zep 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 Zep 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 Zep 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 Zep, you can reuse the same structure for other MCP-enabled apps with minimal code changes.

How to build Zep MCP Agent with another framework

OpenAI Agents SDK

OpenAI Agents SDK

Use Zep MCP with OpenAI Agents SDK

Claude Agent SDK

Claude Agent SDK

Use Zep MCP with Claude Agent SDK

Google ADK

Google ADK

Use Zep MCP with Google ADK

LangChain

LangChain

Use Zep MCP with LangChain

Vercel AI SDK

Vercel AI SDK

Use Zep MCP with Vercel AI SDK

Mastra AI

Mastra AI

Use Zep MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Zep MCP with LlamaIndex

CrewAI

CrewAI

Use Zep MCP with CrewAI

Claude Code

Claude Code

Use Zep MCP with Claude Code

Codex

Codex

Use Zep MCP with Codex

FAQ

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

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

Can I use Tool Router MCP with Autogen?

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 Zep tools.

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

Yes, absolutely. You can configure which Zep 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 Zep 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