🦇 Vigilant-AI - The Ultimate Misinformation & Deepfake Detector

Vigilant-AI is an advanced fact-checking and AI-powered image verification tool designed to combat misinformation. It includes:

• A web app for checking facts and detecting AI-generated images
• A Chrome extension for right-click verification on any webpage

✨ Features Overview

✅ Fact Checking with Reliable Sources

• What it does: Uses Falcon-7B (Hugging Face) to analyze textual claims for credibility.
• Why we chose this: To fight misinformation and provide users with quick verification.
• How to use:
  1. Enter a claim in the text box.
  2. Click "Check Fact".
  3. View AI-generated analysis + Wikipedia & Google News references.

---

✅ AI Image Fake Detection (Upload & URL-based)

• What it does: Uses EfficientNetB0 to analyze images and determine whether they are AI-generated or real.
• Why we chose this: Deepfake images and AI-generated visuals are increasing rapidly, making verification crucial.
• How to use:
  1. Upload an image (or drag & drop).
  2. Click "Analyze Image".
  3. Alternatively, right-click any image on a website and choose "Check Image with Vigilant-AI".
  4. The tool provides a confidence score on authenticity.

---

✅ History Tracking for Fact & Image Checks

• What it does: Stores up to 10 previous verifications for easy reference.
• Why we chose this: Users may want to revisit past verifications.
• How to use:
  • View past verifications in the History section.
  • Click "Clear History" to reset.

---

✅ Drag & Drop Image Upload for Faster Analysis

• What it does: Allows users to drag & drop images for analysis.
• Why we chose this: Enhances usability & accessibility.
• How to use:
  • Drag an image into the drop zone.
  • It auto-uploads and previews the image before analysis.

---

✅ Right-Click Image URL Checking (Chrome Extension)

• What it does: Enables users to verify images directly from webpages.
• Why we chose this: Many fake images spread online, and this allows instant verification.
• How to use:
  • Right-click an image on any webpage.
  • Select "Check Image with Vigilant-AI".
  • The extension will analyze and display results in a pop-up.

---

🛠 How It Works (Tech Stack & Architecture)

Frontend

• Built with HTML, CSS, JavaScript
• Includes a Chrome extension

Backend

• Python + Flask API for processing requests
• AI models:
  • Falcon-7B (Hugging Face API) for fact verification
  • EfficientNetB0 for image analysis

Storage & History

• Uses LocalStorage to store recent fact-checks & analyses

---

💻 How to Run the Project Locally

1️⃣ Clone the Repository

$ git clone https://github.com/pranjal-coders/Vigilant-AI.git
$ cd Vigilant-AI

2️⃣ Install Dependencies

$ pip install -r requirements.txt

📌 Ensure Python 3.8+ is installed before running the setup.

3️⃣ Set Up API Keys

1. Hugging Face API Key (for fact-checking):
   • Go to Hugging Face.
   • Generate an API key with read & write access.
   • Create a .env file in the project folder and add:

     HUGGINGFACE_API_KEY=your_api_key_here

4️⃣ Start the Backend Servers

$ python main.py  # Starts fact-checker (Port 5000)
$ python app.py   # Starts image analyzer (Port 5001)

5️⃣ Open the Web App

Visit http://127.0.0.1:5000/ in your browser.

6️⃣ Install the Chrome Extension

1. Open chrome://extensions/
2. Enable Developer Mode (top right corner).
3. Click "Load unpacked" and select the /extension folder.
4. The extension is now installed & ready to use!

---

📈 Contribution Guidelines

We welcome pull requests and contributions!

How to Contribute:

1. Fork the repository
2. Create a new branch for your feature
3. Make changes & commit
4. Submit a pull request with a clear description

Ideas for Contributions:

✔ Improve fact-checking model accuracy 

✔ Add real-time news sources (e.g., Google Fact Check API)

✔ Expand image verification to detect deepfakes

✔ Optimize performance & UI

---

✨ Future Improvements

• Enhance AI model accuracy for fact verification.
• Deploy backend to a public cloud server (so users don’t need to run Flask locally).
• Support multiple languages for global accessibility.
• Add deepfake video detection in future versions.

---

🔒 License

This project is licensed under the MIT License.

---

💙 Made with Love by the Vigilant-AI Team!

For any questions, feel free to open an issue on GitHub! 🚀