- Overview
- Features
- Architecture
- Installation
- Running the Application
- Usage
- How It Works
- Contributing
- License
- Acknowledgements
This project demonstrates an intelligent travel planning system built with Google Agent Development Kit (ADK). It showcases a multi-agent architecture where specialized AI agents collaborate to create comprehensive travel recommendations for users.
The system leverages Gemini 2.0 Flash for intelligent decision-making across multiple specialized agents, all coordinated through a host agent and presented through an intuitive Streamlit interface.
|
Specialized AI agents working together to create comprehensive travel plans |
Leveraging Gemini 2.0 Flash model for intelligent travel suggestions |
|
Simple Streamlit interface for travel planning inputs |
Coordinated planning through API calls between specialized agents |
The project consists of the following components:
| Agent | Port | Description |
|---|---|---|
| 🎮 Host Agent | 8000 | Acts as the coordinator, distributes requests to specialized agents, and aggregates responses into a unified travel plan |
| 8001 | Recommends flight options based on origin, destination, dates, and budget, providing airline, price, and timing details | |
| 🏨 Stay Agent | 8002 | Suggests accommodations at the destination with hotel names, pricing, and locations |
| 🎭 Activities Agent | 8003 | Recommends tourist and cultural activities with descriptions, prices, and durations |
The frontend is built with Streamlit, providing an intuitive interface for users to input their travel details and view the comprehensive recommendations.
- Python 3.7+
- Git
- OpenAI/Gemini API key
1. Clone the Repository
[email protected]:bright-arparwut/traveling-planer-agents-A2A.git
cd traveling-planer-agents-A2A2. Set up Python Environment
# Create and activate virtual environment
python3 -m venv adk_demo
source adk_demo/bin/activate # On Windows: adk_demo\Scripts\activate
# Install dependencies
pip install -r requirements.txt3. Configure API Key
# Linux/Mac
export OPENAI_API_KEY="your-api-key"
# Windows
set OPENAI_API_KEY=your-api-keyStart each component in separate terminal windows:
Launch All Components
# 1. Start Host Agent
uvicorn agents.host_agent.__main__:app --port 8000 &
# 2. Start Flight Agent
uvicorn agents.flight_agent.__main__:app --port 8001 &
# 3. Start Stay Agent
uvicorn agents.stay_agent.__main__:app --port 8002 &
# 4. Start Activities Agent
uvicorn agents.activities_agent.__main__:app --port 8003 &
# 5. Launch Streamlit UI
streamlit run travel_ui.pyThe Streamlit interface will automatically open in your browser at http://localhost:8501.
-
Enter Travel Details:
- Origin city (e.g., "Bangkok")
- Destination city (e.g., "Paris")
- Start date
- End date
- Budget in USD
-
Click "Plan My Trip!"
-
View Results:
- Flight recommendations
- Accommodation options
- Suggested activities
- All within your specified budget!
- User submits travel details through the Streamlit UI
- Request is sent to the Host Agent (port 8000)
- Host Agent distributes the request to specialized agents:
- Flight Agent (port 8001)
- Stay Agent (port 8002)
- Activities Agent (port 8003)
- Each specialized agent uses the Gemini API to generate relevant suggestions
- Host Agent aggregates all responses
- UI displays the complete travel plan to the user
Contributions are welcome! Here's how you can help:
- Fork the repository
- Create a feature branch:
git checkout -b new-feature - Commit your changes:
git commit -am 'Add new feature' - Push to the branch:
git push origin new-feature - Submit a pull request
This project is licensed under the MIT License - see the LICENSE file for details.
- Google Agent Development Kit (ADK) for the agent framework
- Gemini AI for the powerful language model capabilities
- Streamlit for the user interface components
Built with ❤️ using Google Agent Development Kit