The AI That Runs Compliance for the World

RegulationOS is a demo-ready web application that uses agentic AI to read regulatory documents, analyze operational data (e.g. cargo manifests), detect compliance risks, and generate actionable remediation plans. The primary demo vertical is Shipping and Transportation: upload a regulation PDF and a cargo manifest CSV to get a compliance score, risk list, and recommended actions.

• Frontend: Next.js 14, TypeScript, Tailwind CSS
• Backend: FastAPI, Python 3.11
• AI: Amazon Nova–ready multi-agent architecture with a mock adapter for offline/demo use (no API key required).

┌─────────────────────────────────────────────────────────────────┐
│                     FRONTEND (Next.js 14)                        │
│   Home │ Upload (PDF + CSV) │ Dashboard (score, risks, Q&A)      │
│                        lib/api.ts → REST                         │
└────────────────────────────┬────────────────────────────────────┘
                              │ HTTP
┌────────────────────────────▼────────────────────────────────────┐
│                      BACKEND (FastAPI)                           │
│   GET /health   POST /analyze   POST /analyze-demo   POST /ask   │
│                              │                                   │
│   SupervisorAgent → RegulationAgent → CargoAgent → RiskAgent     │
│                              → WorkflowAgent   VoiceAgent (/ask)  │
│   NovaClient / MockNovaClient   StorageService (JSON reports)   │
└─────────────────────────────────────────────────────────────────┘

Agents

1. Clone and enter the project

   cd regulation-os

2. Environment

   cp .env.example .env

   Edit .env if you want to use the live Nova API; for demo, USE_MOCK_NOVA=true is enough.

3. Backend (from project root regulation-os/)

   python -m venv .venv
   .venv\Scripts\activate   # Windows
   # source .venv/bin/activate   # macOS/Linux
   pip install -r backend/requirements.txt
   set PYTHONPATH=%CD%   # Windows
   uvicorn backend.main:app --reload --port 8000

4. Frontend

   cd frontend
   npm install
   npm run dev

   Open http://localhost:3000.

From the regulation-os directory:

cp .env.example .env
docker-compose up --build

• Backend: http://localhost:8000
• Frontend: http://localhost:3000
• API docs: http://localhost:8000/docs

• With USE_MOCK_NOVA=true (default), the app uses MockNovaClient: deterministic, offline, no API key.
• To use the live Nova API, set USE_MOCK_NOVA=false and NOVA_API_KEY in .env. The same agent code then calls the real API.

1. Open http://localhost:3000.
2. Click “Launch Demo” (or go to Upload).
3. Click “Load Demo Data” to load the sample regulation and cargo manifest.
4. Click “Analyze”. Wait for the run to finish.
5. You are redirected to the Dashboard: compliance score, risk list, recommended actions, workflow automations.
6. In Ask RegulationOS, type a question (e.g. “What is the main compliance risk?”) and click Ask to see the Q&A response.

• Backend: From regulation-os, with PYTHONPATH set to the project root:

  pytest backend -v

• Frontend: From regulation-os/frontend:

  npm run test

• Persist report IDs and support “load report by ID” from the dashboard.
• Add more sectors (Healthcare, Finance, Energy) with sector-specific regulations and schemas.
• Replace mock with full Nova integration (embeddings, Sonic) for production.
• Optional auth and tenant isolation for multi-tenant deployments.