Skip to main content
Google ADK and Klavis Integration - Build AI agents with MCP tools
You can find the complete example code in Klavis GitHub repository here ->

​
Prerequisites

Before we begin, you’ll need: Google API key and Klavis API key.

​
Installation

First, install the required packages: 
pip install google-adk klavis

​
Setup Environment Variables

import os
from dotenv import load_dotenv

load_dotenv()

os.environ["KLAVIS_API_KEY"] = "YOUR_KLAVIS_API_KEY"  # Replace

​
Step 1 - Create an Agent Project

Run the ADK create command to start a new agent project: 
adk create my_agent
This will create the following project structure: 
my_agent/
    agent.py      # main agent code
    .env          # API keys or project IDs
    __init__.py

​
Step 2 - Configure Agent with Klavis MCP

The agent.py file contains a root_agent definition which is the only required element of an ADK agent. Update your agent.py to integrate Klavis MCP servers: 
import os
import webbrowser

from google.adk.agents.llm_agent import Agent
from google.adk.tools.mcp_tool import StreamableHTTPConnectionParams
from google.adk.tools.mcp_tool.mcp_toolset import McpToolset
from klavis import Klavis
from klavis.types import McpServerName

from dotenv import load_dotenv
load_dotenv()

KLAVIS_API_KEY = os.getenv("KLAVIS_API_KEY")

# Initialize Klavis and set up Strata server
klavis_client = Klavis(api_key=KLAVIS_API_KEY)

user_id = "user_123"

# Create Strata server with multiple MCP servers
strata_response = klavis_client.mcp_server.create_strata_server(
    servers=[McpServerName.GMAIL, McpServerName.SLACK],
    user_id=user_id
)

# Handle OAuth authentication
if strata_response.oauth_urls:
    for server_name, oauth_url in strata_response.oauth_urls.items():
        user_integration_auth = klavis_client.user.get_user_auth(
            user_id=user_id,
            server_name=server_name
        )
        if not user_integration_auth.is_authenticated:
            print(f"🔐 Opening OAuth for {server_name}...")
            webbrowser.open(oauth_url)
            input(f"Press Enter after completing {server_name} OAuth authorization...")

mcp_server_url = strata_response.strata_server_url

# Create AI agent with MCP toolset (exposed at module level for ADK)
root_agent = Agent(
    name="my_agent",
    model="gemini-2.5-flash",
    description="An agent with access to tools through Klavis MCP",
    instruction="You are a helpful assistant with access to MCP tools.",
    tools=[
        McpToolset(
            connection_params=StreamableHTTPConnectionParams(
                url=mcp_server_url,
            ),
        )
    ],
)
OAuth Authorization Required: The code above will open browser windows for each service. Click through the OAuth flow to authorize access to your accounts.

​
Step 3 - Run Your Agent

Launch the web interface to interact with your agent: 
adk web
This will start a local web server where you can chat with your agent and watch it use the Gmail and Slack MCP tools.
Perfect! You’ve integrated Google ADK with Klavis MCP servers.

​
Next Steps

Integrations

Explore available MCP servers

API Reference

REST endpoints and schemas

​
Useful Resources

Happy building! 🚀