Product Ingredient Agent 🔍

Multimodal Agentic Workflow - Phidata

A Streamlit application that analyzes product ingredients using Gemini AI to provide insights about food and personal care products.

Features 🌟

• Example Products: Pre-loaded examples of common products
• Image Upload: Upload your own product images
• Camera Capture: Take photos directly through the app
• AI Analysis: Powered by Google's Gemini 2.0 Flash and Tavily Search
• Ingredient Insights: Get a detailed analysis of ingredients and their implications

Demo 🎥

<Tutorial will be released on YouTube. Meanwhile, watch Agentic RAG using Phidata >

Installation 🚀

1. Clone the repository:

git clone https://github.com/yourusername/Product-Ingredient-Agent.git
cd Product-Ingredient-Agent

2. Create a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

Configuration ⚙️

1. Create a .env file in the project root:

TAVILY_API_KEY = your_tavily_api_key
GOOGLE_API_KEY = your_gemini_api_key 

2. Add your example images in the images/ directory:

images/
├── hide_and_seek.jpg
├── bournvita.jpg
├── lays.jpg
└── shampoo.jpg

Usage 💡

1. Run the Streamlit app:

streamlit run app.py

2. Open your browser and navigate to http://localhost:8501

3. Choose one of three options to analyze a product:

   • Select from example products
   • Upload your own image
   • Take a photo using your camera

Project Structure 📁

product-ingredient-analyzer/
├── app.py                 # Main Streamlit application
├── constants.py           # System prompts and constants
├── requirements.txt       # Project dependencies
├── images/               # Example product images
└── README.md             # Project documentation

Dependencies 📚

• streamlit
• phidata
• pillow
• tavily-python
• google-generativeai

Contributing 🤝

⭐️ STAR the Phidata repository: https://github.com/phidatahq/phidata

License 📄

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments 👏

• Phidata for building Multimodal Agent
• Google Gemini AI for powering the analysis
• Streamlit for the web interface
• Tavily for search capabilities