Inbox Eagle

An app that queries your unread emails and ranks them in importance based on which ones need a reply from you.

Configurable to use either a local LLM or an OpenAI API model.

Check out this article for a walkthrough of the application in more depth.

Getting started

1. cp .env.example .env
2. cp prompt-config.example.json prompt-config-setup.json
3. Set up Gmail API credentials
   • You can create access credentials via the Gmail Cloud console API page
   • Enable the following scope: https://www.googleapis.com/auth/gmail.modify
   • Set the application type to "Web Application"
   • Add an Authorized Redirect URI of: http://localhost:33339/
   • Download the Gmail credentials JSON file
4. Place the Gmail credentials JSON in this directory named credentials-gmail.json.
5. pip install -r requirements.txt
6. cd client && npm install

Running the script

Run the command line script, which authorizes Gmail and runs the evaluation process:

$ python run_script.py

This will save your Gmail credentials to a token.pickle file locally.

Run the server:

$ python start.py

Run the front-end client:

$ cd client
$ npm start

Screenshots

Web UI:

Output from run_script.py command line script:

-------- Evaluating messages (2) --------

Message ID: 18d71921aacafc7a - Subject: ['Your utilities statement is overdue']
Snippet: Please submit payment for $100.00 for your electric bill. Kindly, City Utilities
Evaluation: This appears to be a payment that needs to be submitted. Therefore, this is given an importance level of 8.0.
Rating: 8.0

----------------

Message ID: 18d659d34ec34e35 - Subject: ['Boeing’s AI Wings It']
Snippet: Plus: Microsoft's AI Explains Itself; Apple's Sixth Sense
Rating: 2.0
Evaluation: Interesting patent. No urgent response necessary. Importance Level: 2

Architecture

• Web frontend
  • User clicks Evaluate emails to contact the Python server
• Python server
  • ✉️ Fetch emails via the Gmail API
  • 🔁 For each email:
    • ✍️ Compose a prompt to the LLM based on the user's personalized configuration
    • 🧠 Query the LLM (twice)
      • First, we ask for a written evaluation of the email's importance, including a score on a scale of 1 to 10
      • Second, we ask the LLM to extract the numeric score from the written evaluation
• Web frontend
  • Display the results of the evaluation!