🖼️ LeetCode Screenshot OCR Solver

This project automates the process of extracting, solving, and explaining LeetCode-style coding problems from desktop screenshots using OCR and OpenAI GPT models.

---

📌 Features

• ✅ Detects open desktop windows and takes a screenshot of a selected one
• 🖼️ Stores screenshots in a dedicated screen_shots/ folder
• 🔍 Uses Tesseract OCR to extract text from screenshots
• 📝 Saves parsed text into the parsed_texts/ folder
• 🤖 Sends the extracted text to OpenAI GPT-4o via API to:
  • Identify any LeetCode-style problem
  • Explain the problem in natural language
  • Provide a step-by-step solution with code
• 📄 Saves the GPT-generated solution in Markdown format under explained_solns/

---

🧠 Example Use Case

Imagine you have a coding challenge open on your screen, and you want:

1. To extract the question text from a screenshot
2. To get an immediate explanation and solution from GPT

This project automates that entire flow.

---

📁 Directory Structure

. ├── screen_shots/ # Screenshots of captured windows ├── parsed_texts/ # Raw OCR text extracted from screenshots ├── explained_solns/ # Markdown solutions from OpenAI GPT ├── .env # Your OpenAI API key (OPENAI_API_KEY="sk-...") ├── notebook.ipynb # The Jupyter Notebook to run everything ├── requirements.txt # Dependency list └── README.md # You're here :)

---

🛠️ Requirements

Make sure the following tools and Python packages are installed:

• Tesseract OCR
• Python 3.8+
• Install dependencies:

pip install -r requirements.txt