PhishGuard ML

• 

  PhishGuard ML — AI-powered multi-channel phishing detection platform

• 

  Screenshot Threat Scanner — Analyze suspicious screenshots for phishing indicators

• 

  Threat Analysis — Detailed security signals and phishing indicators

Inspiration

Phishing attacks are becoming more common and can appear through suspicious URLs, emails, messages, and fake webpages. We wanted to build a simple security tool that helps users identify potential phishing threats without requiring technical cybersecurity knowledge.

What it does

PhishGuard ML is a web-based phishing detection platform that analyzes multiple types of content:

• URL Detection
• Email Detection
• Message Detection
• Screenshot Detection

Users can submit suspicious content and receive a phishing detection result through a simple web interface. The platform also provides authentication, dashboard, and detection history features.

How we built it

We built PhishGuard ML using a React and Vite frontend connected to a Python FastAPI backend.

Technology Stack

• React
• Vite
• Tailwind CSS
• Framer Motion
• Axios
• React Router
• Python
• FastAPI
• Machine Learning
• SQLite
• JWT Authentication

The frontend is deployed on Vercel and the backend is deployed on Render. The frontend communicates with the backend through REST APIs.

Challenges we ran into

One of our main challenges was connecting the frontend with the machine learning backend and handling API communication correctly.

We also worked on authentication, CORS configuration, deployment, error handling, and making sure the different detection modules work through a single web application.

Accomplishments that we're proud of

We are proud of building a complete full-stack phishing detection platform with multiple detection methods in one application.

We successfully connected the React frontend with the FastAPI backend and deployed the application online so that users can access the working prototype.

What we learned

Through this project, we learned how to build and connect a full-stack application, integrate machine learning with a web application, work with REST APIs, implement authentication, handle deployment, and test different types of phishing detection workflows.

What's next for PhishGuard ML

We plan to improve the detection models, expand the types of phishing threats that can be analyzed, improve detection accuracy, add more cybersecurity insights, and make the platform more useful for everyday users.

Built With

• axios
• fastapi
• framermotion
• javascript
• jwt
• machine-learning
• python
• react
• react-router
• recharts
• render
• sqlite
• tailwind
• vercel
• vite