Mind Mirror Journal

Talk to a slightly-evolved version of yourself through a journal, helping you and guiding you through your thoughts <3

Comment

Inspiration

“What if we could talk to a wiser version of ourselves when we’re overwhelmed?” As someone who journals for emotional clarity, I oftentimes wish I had someone to redirect my emotions and guide me through them. This is what sparked Mind Mirror!

What it does

Mind Mirror is a journal-style web app where you write your thoughts, and an AI version of you (kinder/bolder/more rational) responds.

You choose a reflection mode: Kinder Me : for emotional comfort and support Bolder Me : for confident, action-oriented guidance Rational Me : for calm, logical reasoning

After writing your journal entry, the AI reads it and replies in that reflection style, helping you process thoughts with more clarity and self-awareness. The AI response appears slowly, like handwriting, on a scrapbook-style page, making it feel intimate and personal.

How I built it

Framework & Backend:

  1. Built with Flask (Python) for routing and logic.
  2. Used openai Python package to interact with the GPT-4.1 API.
  3. Managed user input and reflective response using Flask session.

Frontend & Design:

  1. Styled using custom CSS with a nostalgic scrapbook aesthetic.
  2. Integrated Google Fonts: Over the Rainbow for journal input. Annie Use Your Telescope for buttons. La Belle Aurore for the AI response.
  3. Designed 3 routes: /Home page with description. /journal for entry input and reflection style selection. /reflection for animated AI response.

AI Logic:

Used predefined personas:

  1. Kind: comforting and supportive
  2. Bold: confident and action-oriented
  3. Rational: logical and calm

Injected these as system prompts into the OpenAI chat request.

Challenges I ran into

  1. Initially tried to use Streamlit, then I realized it wasn't compatible for the full vision I have for Mind Mirror.
  2. I had to ensure the AI response didn't overflow visually, so I implemented scrollable containers only when the content exceeded a set height.

What I learned

This project taught me to merge aesthetics and AI into a seamless user experience. Specifically, I learned:

  1. How to integrate OpenAI's gpt-4.1 model with a Flask web app.
  2. How to use sessions in Flask to persist state across routes.
  3. I also deepened my understanding of how personality prompts shape AI responses.

Built With

Share this project:

Updates