📧 Chat with Your Gmail Inbox using RAG

An AI-powered Gmail assistant that lets you ask questions about your inbox using natural language. Built with Retrieval-Augmented Generation (RAG) to retrieve relevant email content and generate answers using an LLM.

Built with Python, Streamlit, OpenAI, Embedchain, ChromaDB, and the Gmail API.

---

✨ Features

• 📬 Chat with your Gmail inbox in natural language
• 🔎 Semantic retrieval of relevant email content
• 🤖 AI-generated answers powered by OpenAI models
• ⚡ Fast vector search using ChromaDB
• 🖥️ Interactive UI built with Streamlit
• 🔐 Secure Gmail authentication via OAuth 2.0

Example queries:

• "Summarize my recent emails"
• "Any emails about payments?"
• "Who emailed me today?"
• "Do I have any emails about interviews?"

---

🧠 How It Works

This app follows a standard RAG (Retrieval-Augmented Generation) workflow:

1. Gmail emails are fetched using the Gmail API
2. Emails are converted into vector embeddings using OpenAI
3. Embeddings are stored in a ChromaDB vector database
4. When a user asks a question:
   • Relevant email chunks are retrieved from the vector database
   • The OpenAI LLM generates an answer using the retrieved context as a prompt

---

🏗️ Architecture

User Question
    → Streamlit Interface
    → Embedchain Pipeline
    → ChromaDB Vector Database
    → Relevant Email Context
    → OpenAI LLM
    → Generated Answer

---

🛠️ Tech Stack

Layer	Technology
Frontend	Streamlit
LLM & Embeddings	OpenAI API
RAG Pipeline	Embedchain
Vector Store	ChromaDB
Email Integration	Gmail API
Auth	OAuth 2.0
Language	Python

---

⚙️ Installation

1. Clone the repository:

git clone https://github.com/YOUR_USERNAME/Gmail_RAG_app.git
cd Gmail_RAG_app

2. Install dependencies:

pip install -r requirements.txt

---

🔑 Gmail API Setup

1. Go to the Google Cloud Console
2. Create a new project
3. Enable the Gmail API
4. Create an OAuth Client ID (Desktop App type)
5. Download the credentials file and rename it to credentials.json
6. Place credentials.json in the project root folder

---

▶️ Running the App

streamlit run app.py

A browser window will open where you can:

1. Enter your OpenAI API key
2. Authenticate with your Google account
3. Start chatting with your Gmail inbox

---

📁 Project Structure

Gmail_RAG_app/
├── app.py
├── requirements.txt
├── README.md
└── .gitignore

---

🔒 Security

The following files contain sensitive credentials and must not be committed to GitHub. Add them to your .gitignore:

credentials.json
token.json

---

💡 Future Improvements

• AI-generated daily email summaries
• Automatic action item extraction
• Email topic clustering
• Semantic email search filters
• Slack / Teams integration

---

📜 Disclaimer

This project is intended for educational and demonstration purposes only.