ProjectDocumentation.txt

# Unknown Unknown: Project Documentation

1. Project Overview

**Unknown Unknown** is a mobile-friendly AI chatbot platform for microlearning. Inspired by feed-based content discovery apps, it allows users to browse educational content in a swipable interface — vertically for switching topics, and horizontally for exploring subtopics within each topic. It aims to empower users to learn in fragmented time using interactive, AI-generated knowledge structures.

2. Use Case & User Flow

* Users open the app and scroll through a feed of "topics" (e.g., a book or knowledge theme).
* Each topic contains "subtopics" (e.g., chapters or ideas), which the user swipes through.
* Each subtopic provides a script (text), optional image, and a chatbot interface powered by GPT-4.
* Users can upload images (e.g., a photo of a book cover) or enter a description, and the system generates a new topic and subtopics using AI.

 3. System Architecture

Backend

* **Language**: Python 3
* **Framework**: Flask
* **Authentication**: JWT
* **Database**: SQLite
* **AI Provider**: OpenAI (GPT-4)
* **File Handling**: base64 images

Data Model

* `User`: id, username, email, password\_hash
* `Topic`: id, title, description, cover\_image
* `Subtopic`: id, topic\_id, script, image\_url
* `UserSavedSubtopics`: user\_id, subtopic\_id, created\_at

4. Codebase Overview

| File               | Responsibility                             |
| ------------------ | ------------------------------------------ |
| `app.py`           | Registers blueprints, starts the Flask app |
| `db.py`            | Manages SQLite DB schema and connections   |
| `auth_routes.py`   | Registration and login logic               |
| `topic_routes.py`  | Topic and subtopic listing/saving APIs     |
| `chat_routes.py`   | GPT-4 conversation per subtopic            |
| `upload_routes.py` | Upload image via base64, return file path  |

5. Functional Modules

 5.1 Authentication

* `POST /register`
* `POST /login`

5.2 Topic/Subtopic

* `GET /topics` — all topics
* `GET /subtopics?topic_id=` — all subtopics for a topic
* `POST /save-subtopic` — save to user collection
* `GET /my-subtopics` — fetch saved subtopics

5.3 GPT Chat

* `POST /chat-with-subtopic` — Chat using context from selected subtopic

5.4 Upload

* `POST /upload-image` — Upload base64-encoded image and return image path

6. Future Expansion

* Topic generation via OCR + GPT-4
* Feed optimization and swipe gesture logic (frontend)
* User-generated summaries
* Integration with a persistent cloud DB and user analytics