🔐 Password Strength Analyzer

An ML and GenAI-based application that analyzes password strength, identifies potential weaknesses, and generates stronger password suggestions.

The system combines machine learning-based strength scoring, dictionary-based analysis using the RockYou dataset, and Mistral AI for generating alternative password suggestions.

🚀 Live Demo

Application: https://passwordstrengthanalyzer.onrender.com

Project Presentation: https://www.canva.com/design/DAGqbLW6gy0/s9N9_gviSXHVTyesxpp7RQ/edit

---

📌 Overview

Password Strength Analyzer evaluates a user's password using multiple approaches rather than relying on a single rule-based check.

The application performs:

User Password
      ↓
Password Analysis
      ↓
┌───────────────────────────┐
│ ML-based Strength Scoring │
│ Dictionary Analysis       │
└───────────────────────────┘
      ↓
Weakness Identification
      ↓
AI-generated Suggestions
      ↓
Stronger Password Alternatives

The project was developed to explore the use of machine learning and generative AI in cybersecurity-oriented applications.

---

✨ Features

• 🔢 Machine Learning Strength Scoring Uses a trained machine learning model to evaluate password strength.

• 📖 Dictionary-Based Analysis Supports analysis against the RockYou password dataset to identify commonly used passwords or password patterns.

• 🤖 AI-Generated Suggestions Uses the Mistral AI API to generate alternative password suggestions.

• ⚙️ Flask Backend Provides the backend application and handles password analysis requests.

• 🔐 Environment-Based Configuration API credentials and configuration values are stored through environment variables rather than hard-coded into the application.

• 🧪 Testable Structure The project includes a dedicated tests directory for testing different inputs and application behavior.

---

🧠 How It Works

1. Password Input

The user provides a password through the application interface.

The password is then passed to the backend for analysis.

2. ML-Based Strength Evaluation

The application uses the trained model:

password_strength_model.joblib

to generate a password strength assessment.

The trained model is loaded by the password-strength analysis component in:

backend/ml_models/password_strength_model.py

3. Dictionary Analysis

The application can also use the RockYou dataset as a source of commonly occurring passwords.

This provides an additional check beyond the machine learning model and helps identify passwords that may be commonly used.

4. AI-Based Suggestions

When stronger alternatives are required, the application can use the Mistral AI API to generate password suggestions.

The GenAI functionality is implemented separately in:

backend/ml_models/genai_suggestions.py

This keeps the AI suggestion component separate from the main password-strength analysis logic.

---

🏗️ Project Structure

PasswordStrengthAnalyzer/
│
├── backend/
│   ├── app.py
│   ├── config.py
│   │
│   └── ml_models/
│       ├── password_strength_model.py
│       └── genai_suggestions.py
│
├── frontend/
│
├── tests/
│
├── password_strength_model.joblib
├── requirements.txt
└── README.md

Main Components

Component	Purpose
app.py	Flask application entry point
config.py	Application configuration
password_strength_model.py	Loads and uses the trained ML model
genai_suggestions.py	Handles Mistral AI-based suggestions
password_strength_model.joblib	Saved machine learning model
tests/	Test cases and mock inputs
frontend/	User interface

---

🛠️ Technology Stack

Backend

• Python
• Flask

Machine Learning

• Scikit-learn / trained ML model
• Joblib

Generative AI

• Mistral AI API

Data

• RockYou password dataset

Testing

• Python testing tools

---

⚙️ Setup

1. Clone the Repository

git clone https://github.com/Harshal0308/PasswordStrengthAnalyzer.git
cd PasswordStrengthAnalyzer

2. Create a Virtual Environment

python -m venv venv

On Windows:

venv\Scripts\activate

3. Install Dependencies

pip install -r backend/requirements.txt

4. Configure Environment Variables

Create a .env file according to the application's configuration requirements.

Example:

MISTRAL_API_KEY=your_mistral_api_key
SSL_CERT_PATH=path_to_certificate

Do not commit .env files or API keys to the repository.

---

🔑 RockYou Dataset

The project supports dictionary-based analysis using the RockYou password dataset.

If the dataset is unavailable, the dictionary-based analysis component cannot perform its intended comparison.

The dataset should be obtained and used according to its applicable terms and restrictions rather than being unnecessarily committed to the repository.

---

🧪 Testing

The project contains a dedicated tests/ directory for testing password inputs and application behavior.

Example test categories include:

• Weak passwords
• Strong passwords
• Common passwords
• Different password patterns
• Mock inputs for application components

Additional test cases can be added as the project evolves.

---

🔒 Security Considerations

This project is intended as an educational and development project for exploring password analysis techniques.

It should not be treated as a complete password-security solution.

In particular:

• Passwords should not be stored unnecessarily.
• API keys must be kept outside the source code.
• Real user passwords should not be included in datasets or test files.
• Password-strength results should be treated as an estimation rather than a guarantee of security.

---

🔮 Future Improvements

• Add more password-analysis features and test cases.
• Improve the machine learning model using a larger and more diverse dataset.
• Provide more detailed explanations for detected password weaknesses.
• Improve the user interface and visualization of strength results.
• Add additional evaluation metrics for the ML model.

---

👥 Contributors

• Sidharth Maharana
• Harshal Kale
• Shiv Tangloo

---

📄 Project Purpose

This project was developed to explore the practical application of machine learning, dictionary-based analysis, Flask, and generative AI in a cybersecurity-related problem.

It demonstrates how multiple analysis techniques can be combined to provide password-strength feedback and generate alternative suggestions.