Inspiration
In the age of social media, misinformation about Islamic teachings spreads rapidly through images, videos, and text posts. We've seen countless instances of fabricated hadiths and misattributed quotes going viral, causing confusion in the Muslim community and a misrepresentation of Islam to non-Muslims. We wanted to create a tool that empowers anyone to instantly verify Islamic content against authentic sources - right from their browser.
The inspiration came from seeing family members share questionable hadith images on WhatsApp and Instagram without knowing if they were authentic. We thought: what if verification was as easy as a right-click?
What it does
Sunnah Fact Checker is a Chrome extension that verifies Islamic claims against authentic hadith sources in real-time. It:
- Extracts text from images and videos
- Analyzes the content using AI-powered fact-checking models
- Searches for authentic hadiths from verified sources like Sahih Bukhari and Sahih Muslim
- Provides instant verification with confidence scores and source references
- Works seamlessly in the browser - just right-click any image or youtube video
Users can verify content in three ways:
- Right-click any image/video on a webpage
- Use the extension popup to enter URLs
The system responds with a verdict (verified/contradicts/unverifiable), confidence level, and direct references to authentic hadith collections.
How we built it
We built Sunnah Fact Checker as a full-stack application with three main components:
Frontend: A Chrome browser extension that seamlessly integrates into users' browsing experience, allowing them to verify content with a simple right-click.
Backend: A Python-based server that handles image and video processing, connects to our hadith database, and performs AI-powered fact-checking using modern machine learning models.
Database: We collected and organized thousands of authentic hadiths from verified Islamic sources into a searchable database that powers our verification system.
The Workflow: When a user submits content, our system extracts any text present, searches our hadith database for relevant sources, and uses AI to compare the claim against authentic references. Results are returned in seconds with confidence scores and citations.
Key Technologies: We leveraged industry-standard tools including Python for backend processing, JavaScript for the browser extension, cloud databases for scalability, and open-source AI models for natural language understanding. Everything is containerized for easy deployment and runs efficiently in the cloud.
The entire system was designed with user privacy in mind - no personal data is collected or stored.
Accomplishments that we're proud of
✨ Built a complete end-to-end pipeline from browser to AI to database in a short timeframe
✨ Achieved real-time verification - results return in 2-5 seconds for most queries
✨ Created a respectful scraper that populated our database with 10,000+ authentic hadiths while honoring source website rate limits
✨ Implemented hybrid search combining traditional text search with semantic similarity for better recall
✨ Made it production-ready with Docker containerization, proper authentication, and comprehensive documentation
✨ User experience - incredibly simple to use (just right-click), no complex setup for end users
What we learned
** We learned valuable skills in fullstack development and AI integration. We learend the caveats of false information on the web and how to navigate around it.
Long-term Vision: Create a comprehensive Islamic knowledge verification platform that:
- Covers Quran, Hadith, and scholarly opinions
- Integrates with Islamic education platforms
- Provides API access for other developers
- Builds a community-driven verification system
The Challenges we faces
** We face challenges in setting up the pipeline through docker. We also found challenges integrating the AI model to check for matching hadiths. **
We envision this becoming a trusted resource that helps preserve authentic Islamic knowledge in the digital age while combating misinformation.
Log in or sign up for Devpost to join the conversation.