AI Email Responder

AI Email Responder is a Streamlit-based application that uses OpenAI's GPT-4o-mini model to automatically generate professional email replies. It helps users save time, maintain a consistent tone, and improve productivity by drafting polite and context-aware responses.

🚀 Features

• Automatically generates email subject and body from raw email content.
• Produces structured JSON output including:
  • reply_subject
  • reply_body
  • tone (Formal | Friendly | Casual)
  • actions (e.g., apologize, request info, escalate)
• Copy-and-send ready email drafts.
• Configurable OpenAI API key via Streamlit secrets or environment variables.
• New Features:
  • File upload for email content (.txt or .eml files)
  • Reply history with save and delete options
  • Copy generated replies to clipboard
  • Improved input validation
  • Support for multiple AI models (GPT-4o-mini, GPT-3.5-turbo)

🛠 Tech Stack

• Frontend: Streamlit
• Backend: Python
• AI Model: OpenAI GPT-4o-mini / GPT-3.5-turbo
• Key Modules:
  • streamlit_app.py – User interface and app flow
  • utils/llm_handler.py – Handles OpenAI API calls
  • utils/email_utils.py – Email parsing and utilities

⚡ Installation

1. Clone the repository:

git clone https://github.com/upstackpilot0710/inbox-ai-agent.git
cd inbox-ai-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

4. Set up OpenAI API key:

   • Create a .streamlit/secrets.toml file in the project root:

     OPENAI_API_KEY = "your-api-key-here"
     

   • Or set as environment variable: export OPENAI_API_KEY=your-api-key-here

5. Run the application:

streamlit run streamlit_app.py

📖 Usage

1. Open the app in your browser.
2. Choose input method: Manual input or upload an email file.
3. Fill in the email details or upload a file.
4. Select preferred tone and signature options.
5. Click "Generate Reply" to get an AI-generated response.
6. Edit the reply if needed, copy to clipboard, or save to history.
7. Simulate sending the reply.

🏗 Workflow / Architecture

User Input → Streamlit UI → LLM Handler → OpenAI API → JSON Output → Display Draft

🧭 Architecture Diagram

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

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

📞 Support

If you have any questions or issues, please open an issue on GitHub.