Bhavesh716/SACH-AI ? reverse-engineered prompt

Build me a hackathon style web app called SACH AI that helps people check if digital content is real or suspicious before they share it.

I want users to be able to upload or paste text, images, screenshots, audio, and videos, then get back a simple result like a trust score, risk level, and a short explanation of why it looks suspicious. If it is an image, try to detect AI generated or edited content and extract any text from it. If it is a video, analyze frames for deepfake signs and also check any audio. If it is audio, transcribe it and look for cloned voice or scam language. If it is text, flag phishing, fake alerts, and misleading claims, and use trusted sources where possible to support the result.

Make the interface clean and easy to use, with an upload area, results page, and clear warnings. Use the README as the guide, and look up current docs online if you need to.