This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine
What I Built
GhostProof ā an AI-powered fact checker and code reviewer.
It verifies factual claims against live web evidence and reviews codebases for security risks, hardcoded credentials, and potential flaws. It brings two essential developer tools into one platform.
Demo
š Live Demo
Code
š» GitHub Repository
How I Built It
Built with React, TypeScript, and Vercel serverless APIs.
- Gemma 4 ā primary AI engine for fact-checking and code reviews.
- Gemini 2.5 Flash ā fallback when the primary model is unavailable.
- JSZip ā browser-side ZIP inspection.
- Deterministic scanning ā detects and masks potential secrets before AI analysis.
- Web evidence search ā provides sources for fact-checking claims.
Why Does Open Innovation Matter?
Open-weight models like Gemma make AI experimentation more accessible and give developers greater flexibility over how AI is integrated into real-world tools. GhostProof is my exploration of putting open AI to work beyond simple chat interfaces.
Prize Categories
- Gemma ā using Gemma 4 as the primary model for both fact-checking and code review.
Built for Hacktoberfest Week 4. š»
Top comments (0)