by Smitha Kolan

This directory contains a sample ADK agent that builds personalized food tours using the Gemma 4 31b model and Google Maps MCP. Given a dish photo or text description, a location, and an optional budget, the agent identifies the cuisine, finds relevant places, creates a walking route, and recommends what to order at each stop.

• Enable Google Maps API on Google Cloud Console.
• Create a Google Maps Platform API key.
• Create a Google AI Studio API key in Google AI Studio.
• ADK installed and configured in your Python environment

1. Clone the sample repository

   git clone https://github.com/google/adk-samples.git

2. Navigate to the sample directory

   cd adk-samples/python/agents/gemma-food-tour-guide

3. Create a virtual environment

   python -m venv .venv

4. Activate the virtual environment

   source .venv/bin/activate

5. Install Dependencies

   pip install -r requirements.txt

6. Configure Environment Variables Rename .env.example to .env and fill in your keys:

   • MAPS_API_KEY: Your Google Maps API Key
   • GEMINI_API_KEY: Your API Key from AI Studio

Run the ADK web interface from the root gemma-food-tour-guide folder:

# Ensure you are in the root food_tour_agent directory
adk web

Then follow the link to chat with the agent! Give it an image or description of food, a location, and a budget.

To test out the capabilities of the Food Tour Agent, try pasting one of these prompts into the chat:

• "I want to do a ramen tour in Toronto. My budget is $60 for the day. Give me a walking route for the top 3 spots and tell me what I should order at each."
• "I have this photo of a deep dish pizza [insert image URL]. I want to find the best places for this around Navy Pier in Chicago. Structure a walking tour and tell me what the must-have slice is at each stop."
• "I'm in Downtown Austin looking for an authentic BBQ tour. Let's keep the budget under $100. Build a walking route between 3 highly-rated spots and give me insider tips on the best cuts of meat to get."

This sample shows how to:

• Use Gemma 4 31b with ADK through the Google AI Studio API
• Connect an ADK agent to Google Maps MCP tools
• Use tool calling to search for places and build routes
• Generate grounded, structured food tour recommendations from Text or Image based input

The agent is configured to:

• use search_places to find relevant restaurants, stalls, or cafes
• use compute_routes to create a walking-optimized itinerary
• rely only on exact place_idvalues or lat_lng objects returned by tools when building routes
• avoid inventing addresses or place identifiers that were not returned by the tools

This helps reduce hallucinations and keeps route generation grounded in tool output.