ConformityAI - Automated Security Compliance Copilot

Pass Audits, Not Weekends. The only AI security assistant that cites its sources 100% of the time.

🚀 Overview

ConformityAI is a production-ready, full-stack RAG (Retrieval-Augmented Generation) application designed to automate the painful process of filling out security questionnaires (SOC 2, SIG, HIPAA).

By ingesting your company's security policies, RFCs, and past questionnaires, ConformityAI builds a private Knowledge Base. It then uses state-of-the-art LLMs to answer incoming security questions in real-time, providing verifiable citations for every claim.

✨ Key Features

• Universal Ingestion: Drag-and-drop PDF (pdf-parse) and text processing.
• Vector Knowledge Base: High-dimensional embeddings using pgvector on Supabase.
• Real-Time RAG Engine: Hybrid search (Keyword + Semantic) with OpenAI text-embedding-3-small.
• Live Streaming Answers: Server-Sent Events (SSE) for instant, typewriter-style AI responses.
• Source Citations: Every answer links back to the specific source document and confidence score.
• Premium UI/UX: Dark-mode aesthetic, glassmorphism, bento-grids, and fluid animations (Next.js 14 + Tailwind).
• Enterprise Ready: Role-Based Access Control (RBAC) ready schema, secure document storage.

🛠 Tech Stack

Frontend ("The Face")

• Framework: Next.js 14 (App Router)
• Language: TypeScript
• Styling: Tailwind CSS, Lucide Icons, clsx
• Animations: Custom Tailwind animations (float, shimmer, pulse-slow)

Backend ("The Brain")

• Runtime: Node.js (Express)
• Language: TypeScript
• Database: Supabase (PostgreSQL + pgvector)
• AI Orchestration: OpenAI API (GPT-4o / GPT-3.5-turbo)
• Storage: Supabase Storage
• Processing: Multer, PDF-Parse

📦 Installation & Setup

Prerequisites

• Node.js v18+
• NPM or Yarn
• A Supabase Account
• An OpenAI API Key

1. Clone the Repository

git clone https://github.com/your-username/conformity-ai.git
cd conformity-ai

2. Backend Setup

1. Navigate to the backend:

   cd backend
   npm install

2. Configure Environment Variables:

   • Create a .env file in backend/:

   PORT=4000
   SUPABASE_URL=your_supabase_project_url
   SUPABASE_SERVICE_KEY=your_supabase_service_role_key
   OPENAI_API_KEY=sk-your-openai-api-key

3. Database Migration:

   • Copy the contents of backend/supabase_schema.sql and run it in your Supabase SQL Editor to create tables.
   • Copy the contents of backend/supabase_rpc.sql and run it to create the vector search function.

4. Start the Server:

   npm run dev

   Server should be running on http://localhost:4000

3. Frontend Setup

1. Navigate to the frontend:

   cd ../frontend
   npm install

2. Start the Dashboard:

   npm run dev

3. Open http://localhost:3000 to launch the app.

� Deployment Guide

Frontend (Vercel)

The frontend is optimized for Vercel.

1. Push code to GitHub.
2. Import project into Vercel.
3. Select frontend as the Root Directory.
4. The build command next build will be auto-detected.
5. Result: You get a live URL like https://conformity-ai.vercel.app.

Backend (Render / Railway)

The backend is a standard Node.js/Express app.

1. Render: Create a new "Web Service".
2. Connect your GitHub repo.
3. Root Directory: backend.
4. Build Command: npm install && npm run build.
5. Start Command: npm start.
6. Add Environment Variables (SUPABASE_URL, etc) in the Render dashboard.
7. IMPORTANT: Update frontend/src/lib/api.ts with your new Backend URL.

�📂 Project Structure

├── backend/
│   ├── src/
│   │   ├── services/
│   │   │   ├── ai/          # LLM & Embedding logic
│   │   │   ├── ingestion/   # File parsers (PDF)
│   │   │   ├── rag/         # Retriever logic
│   │   │   └── store/       # Vector DB interactions
│   │   ├── routes/          # Express API endpoints
│   │   └── index.ts         # Entry point
│   ├── custom_types/        # TypeScript definitions
│   └── supabase_schema.sql  # Database init script
│
├── frontend/
│   ├── src/
│   │   ├── app/             # Next.js App Router Pages
│   │   ├── components/      # React UI Components
│   │   └── lib/             # API clients & utilities
│   └── tailwind.config.ts   # Custom theme & animations

🛡 Security & Compliance

• Zero Data Retention: The AI model is stateless; it does not train on customer data.
• Encrypted Storage: Documents are stored in secure Supabase buckets.
• Vectorization: Text is converted to standard vector arrays, unreadable without the embedding model.

📄 License

Commercial License. Owned by Gobrand Tech.