Hackathon Book Agent: ADK, MCP Toolbox for Databases and A2A

This sample uses the Agent Development Kit (ADK) to create a root (consuming) agent and a remote agent which communicate with each other via ADK and interact with a Cloud SQL PostgreSQL database via MCP Toolbox for Databases.

Overview

The Book Search A2A sample consists of:

• Root Agent (agent.py): A root agent that uses a tool to search for a book's ISBN and then queries a remote library agent for book information.
• Remote Library Agent (hack_library_agent/agent.py): The remote agent that provides information about book location and number of copies, based on ISBN.

Architecture

┌─────────────────┐    ┌────────────────────┐
│ Root Book Agent │───▶│ Remote A2A Library │
│                 │    │        Agent       │
| (localhost:8000)│    |   (GCR_URL:8001)   │
└─────────────────┘    └────────────────────┘

Key Features

1. A2A and MCP

• The hack_book_agent connects to a remote A2A agent hack_library_agent for library queries.
• Both the hack_book_agent and hack_library_agent connect with a database using MCP Toolbox for Databases.
• Demonstrates distributed agent deployment using A2A architecture and MCP Toolbox for Databases.

2. Uvicorn Server Deployment

• The remote agent is served using uvicorn, a lightweight ASGI server.
• Shows how to expose A2A agents as standalone web services.

3. Agent Functionality

• Book Search: Looks up ISBN codes by book title using a tool.
• Library Query: Uses the ISBN to get location and copy count from the remote agent.

Setup

Same as https://google.github.io/adk-docs/get-started/quickstart/#set-up-environment-install-adk :

Create & Activate Virtual Environment (Recommended):

python -m venv .venv

Activate (each new terminal)

• macOS/Linux: source .venv/bin/activate

• Windows CMD: .venv\Scripts\activate.bat

• Windows PowerShell: .venv\Scripts\Activate.ps1

Install ADK and A2A

pip install google-adk google-adk[a2a]

Setup gcloud environment

touch .env

1. Set up a Google Cloud project and enable the Vertex AI API.
2. Set up the gcloud CLI.
3. Authenticate to Google Cloud from the terminal by running gcloud auth application-default login.
4. When using Python, open the .env file located inside (hack_book_agent/). Copy-paste the following code and update the project ID and location.

GOOGLE_GENAI_USE_VERTEXAI=TRUE
GOOGLE_CLOUD_PROJECT=YOUR_PROJECT_ID
GOOGLE_CLOUD_LOCATION=LOCATION

Run the Agent

Authenticate with Google Cloud. Ensure you have the Vertex AI user role required to use the LLMAgent.

gcloud auth application-default login

Run the Main Agent in the root directory (the remote agent is already running on Cloud Run):

# In a separate terminal, run the adk web server
adk web .

Step 1: Open the URL provided (usually http://localhost:8000 or http://127.0.0.1:8000) directly in your browser.

Step 2. In the top-left corner of the UI, you can select your agent in the dropdown. Select "hack_book_agent".

Example Interactions

Once both services are running, you can interact with the root agent:

Book Location:

User: What is the ISBN code of Norwegian Wood
Bot: The ISBN code of Norwegian Wood is XXX-XXXXXXXXXX

Number of Copies:

User: How many copies of 'No Longer At Ease' are available?
Bot: There are X copies available.

---