Happy slice of mcp pizzaMCP.pizza

adk-mcp-gemma3

MCP.Pizza Chef: arjunprabhulal

adk-mcp-gemma3 is an MCP client that demonstrates building AI agents using Google's Agent Development Kit (ADK), the Model Context Protocol (MCP), and the Gemma 3 model via Ollama. It showcases practical AI workflows such as YouTube search assistance by orchestrating context and tools through MCP, enabling real-time, multi-step reasoning and interaction with external data sources and APIs.

Unmaintained · No commits in 16 months.

Use This MCP client To

Build AI agents combining Google ADK and MCP protocols Implement YouTube search assistant with real-time context Integrate Gemma 3 model for advanced AI reasoning Orchestrate multi-step AI workflows using MCP clients Connect AI agents to external APIs via MCP tooling Demonstrate AI agent capabilities with Ollama integration

README

YouTube Search Assistant with ADK, MCP and Gemma 3

A practical implementation demonstrating YouTube search functionality using Google's Agent Development Kit (ADK), Model Context Protocol (MCP), and the Gemma 3 model via Ollama.

ADK Gemma3 Demo

ADK Gemma Web Demo

📚 Table of Contents

🔍 Introduction

This project showcases how to leverage Google's ADK (Agent Development Kit) and MCP (Model Context Protocol) to build an agent powered by Gemma 3, Google's latest large language model. It demonstrates how to:

  • Connect to locally-hosted Gemma 3 via Ollama
  • Implement YouTube search functionality using MCP
  • Create a conversational agent that can format and present search results

🏗️ Architecture

🧩 Core Components

  • Google ADK - Provides the agent framework
  • Model Context Protocol (MCP) - Standardizes tool communication
  • Gemma 3 (12B) - Powers the language understanding and generation
  • Ollama - Hosts the Gemma model locally
  • MCP YouTube Search - Provides YouTube search capabilities
  • Python 3.9+ - Base runtime environment

adk-mcp-gemma3 FAQ

How does adk-mcp-gemma3 utilize the Gemma 3 model?
It integrates the Gemma 3 model via Ollama to provide advanced AI reasoning within the MCP client framework.
What role does Google ADK play in this MCP client?
Google ADK provides the agent development framework that works with MCP to build and manage AI agents.
Can adk-mcp-gemma3 handle multi-step reasoning tasks?
Yes, it orchestrates multi-step reasoning by managing context and tool interactions through MCP.
Is this client limited to YouTube search functionality?
No, while the demo focuses on YouTube search, the client architecture supports broader AI agent workflows.
How does MCP enhance the capabilities of this client?
MCP standardizes context flow and tool calling, enabling secure, real-time interaction with external data and APIs.
What technologies are integrated in adk-mcp-gemma3?
It integrates Google ADK, MCP, the Gemma 3 model via Ollama, and supports real-time AI agent orchestration.
How can developers get started with adk-mcp-gemma3?
Developers can follow the GitHub repository's getting started guide to set up the environment and run the demo.
Does this client support multiple LLM providers?
While it primarily uses Gemma 3 via Ollama, the MCP framework allows integration with other LLM providers like OpenAI and Claude.