MEOW: Monitoring & Evaluation of Online Writing

Your personal detective in the world of disinformation. Built with ❤️ during EUDIS Defence Hackathon.

Why MEOW? (Our Mission)

The internet is flooded with disinformation, and well-written fake news easily lowers our guard. Manual source-checking takes too much time, and the "red flags" are often invisible to the average user.

We believe that truth and transparency shouldn't be hidden behind a paywall. That's why we created MEOW as a 100% open-source, non-profit project. We want to empower everyday users to critically evaluate online content and help researchers study disinformation.

Core Features

We don't just throw random numbers at the user. MEOW is built on transparency and analyzes websites across three/four main pillars (depends on the website type):

• Domain Trust: Checks SSL security, domain age, and registration history.
• Language Trust: Detects highly emotional tone, slang, or manipulative writing styles.
• Source Trust: Identifies the presence (or suspicious absence) of owners, publication dates, and credible citations.
• Transparency: Checks transparency of news articles etc. by analyzing the authors, date and text structure.

How it Works (The Pipeline)

1. Scraping: Extracts text and metadata directly from the active webpage.
2. Heuristics & AI Analysis: Evaluates technical indicators and uses AI to identify the page type and language patterns.
3. Scoring Engine: Calculates a weighted risk score based on the extracted metrics.
4. Real-time UI: Generates instant signals and warnings in the browser extension.

Installation (For Developers)

Since we are in active development, you can run the MEOW extension locally:

1. Clone this repository:

   git clone https://github.com/[YOUR_USERNAME]/MEOW.git

2. Enter the backend folder:

   cd backend

3. Create a virtual environment:

   python -m venv venv

   Windows:

   venv\Scripts\activate

   Mac/Linux:

   source venv/bin/activate

4. Install dependencies:

   pip install -r requirements.txt

5. Get an API key for gemma-3-4b-it.

6. Create the .env file (in the backend folder).

   .env Structure:

   GEMINI_API_KEY=[YOUR_API_KEY]
   GEMINI_MODEL=gemma-3-4b-it
   PORT=8000

7. Run the server:

   python main.py

8. Install the extension in your browser.

9. Congrats, you can now run the project :3