This tool helps to create a focused environment, highlighting the crucial unread emails by sorting them based on severity and sentiment.
Mail Clipper is a web application developed using the EJS, Passport.js, Express.js, Node.js, and MongoDB stack. It incorporates the gpt-3.5-turbo-1106 model for summarization, sentiment analysis, and severity assessment.
- Google OAuth 2.0 authentication
- Filter emails into categories such as Urgent, Moderate, Routine, and Ignore for easy grouping and organization.
- Date Filter to shortlist emails based on date, displaying summarized emails by default in descending order from the most recent to the last.
- Improve readability by adding an 'Open in Gmail' button so users can easily access the actual email for a better reading experience.
The Classroom Project is built using the following technologies:
- Frontend: EJS, CSS
- Backend: Node.js, Express.js
- Database: MongoDB
- User Authentication: Passport.js
It is recommended to run this project locally to maintain control over critical Gmail data. Follow the below steps:
- Clone the repository:
https://github.com/dublin74/Mail-Clipper.git
- Install dependencies:
npm install
ornpm i
- Create a .env file and update the following credentials created by you
MONGO_DB = "your-mongodb-connection-string"
CLIENT_ID = "your-client-id-here"
ClIENT_SECRET = "your-client-secret-here"
OPENAI_API_KEY = "your-api-key-here"
- Start the server:
node app.js
While creating the client_id and client_secret on the Google Cloud console, remember to use the correct scope for accessing Gmail data
https://www.googleapis.com/auth/gmail.readonly
, and don't forget to put the callback URL; by default, it ishttps://localhost:3000/auth/google/callback
.
After installation, you can access Mail Clipper in your web browser at https://localhost:3000
. Login with your Gmail account, and you are good to go.