Skip to content

Yacinewhatchandcode/EU-AI-Act-Compliance

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Prime-AI Logo

🇪🇺 Prime-AI — EU AI Act Compliance System

The first open-source toolkit to scan, classify, audit, and report AI systems under EU Regulation 2024/1689


🎬 Product Demo

https://github.com/Yacinewhatchandcode/EU-AI-Act-Compliance/raw/main/demo_videos/PRIME_AI_30s.mp4

6 use cases recorded — Landing Page • Auto-Login • Risk Classifier • URL Scanner • 9-Requirement Audit • Knowledge Base


💡 What is this?

Prime-AI is a full-stack compliance toolkit for the EU AI Act (Regulation 2024/1689) — the world's first comprehensive AI regulation, effective August 2, 2026.

Feature Description
🔍 URL Scanner Scan any website to detect AI systems and assess compliance risk
⚖️ Risk Classifier Classify AI systems into 4 levels: Prohibited → High → Limited → Minimal
📋 9-Requirement Audit Full audit against Articles 8-15 with weighted scoring
📊 Compliance Reports Generate reports with remediation roadmaps
📚 Knowledge Base Complete regulatory database — 8 prohibited, 8 high-risk, 9 requirements
🤖 Multi-Platform Web PWA + Telegram + Slack + WhatsApp + Discord

⚡ Quick Start

# Clone
git clone https://github.com/Yacinewhatchandcode/EU-AI-Act-Compliance.git
cd EU-AI-Act-Compliance

# Install (optional — stdlib only, zero mandatory deps)
pip install -r requirements.txt  # only if you want AI-powered analysis

# Run
python eu_ai_act_server.py

# Open → http://localhost:8080

That's it. Zero config required. Auto-login in dev mode. No database. No API keys needed.


🎯 Use Cases — All Recorded as Video

# Use Case Video Duration
1 Marketing Landing Page uc1_landing.mp4 ~20s
2 Zero-Click Auto Login uc2_auto_login.mp4 ~11s
3 AI Risk Classifier uc3_classifier.mp4 ~14s
4 URL Compliance Scanner uc4_scanner.mp4 ~14s
5 9-Requirement Audit uc5_audit.mp4 ~14s
6 Knowledge Base Browse uc6_kb.mp4 ~12s

All demos recorded autonomously using Playwright. See record_demo.py.


🏗️ Architecture

┌─────────────────────────────────────────────────────┐
│              Client Layer (PWA / Bots)              │
│   Web App  ·  Telegram  ·  Slack  ·  WhatsApp  ·  Discord │
└──────────────────────┬──────────────────────────────┘
                       │ HTTP / WebSocket
┌──────────────────────┴──────────────────────────────┐
│              eu_ai_act_server.py (API)              │
│   JWT Auth  ·  REST API  ·  Static Files            │
└──────────────────────┬──────────────────────────────┘
                       │
┌──────────────────────┴──────────────────────────────┐
│              eu_ai_act.py (Core Engine)             │
│   Classifier  ·  Auditor  ·  Scanner  ·  Reporter  │
│   Regulatory DB  ·  Risk Matrix  ·  Remediation     │
└─────────────────────────────────────────────────────┘

Project Structure

EU-AI-Act-Compliance/
├── eu_ai_act_server.py    # HTTP API server + JWT auth
├── eu_ai_act.py           # Core compliance engine
├── bot_engine.py          # Shared bot command brain
├── web/                   # PWA frontend
│   ├── index.html         # Main app (Material Design 3)
│   ├── landing.html       # Marketing landing page
│   ├── login.html         # Auth page (auto-login capable)
│   ├── app.js             # Client-side logic
│   ├── style.css          # Premium dark theme
│   └── manifest.json      # PWA manifest
├── compliance/            # GDPR & EU AI Act compliance docs
├── deploy/                # Docker + VPS deployment configs
├── demo_videos/           # 6 use case recordings + promo
│   ├── PRIME_AI_30s.mp4   # 30s fast marketing video
│   └── PRIME_AI_PROMO.mp4 # 90s full walkthrough
├── telegram_bot.py        # Telegram integration
├── slack_bot.py           # Slack integration
├── discord_bot.py         # Discord integration
└── whatsapp_bot.py        # WhatsApp integration

🔌 API Reference

All endpoints require Authorization: Bearer <token> (auto-generated in dev mode).

Method Endpoint Body Description
GET /api/auth/dev Get dev JWT token
GET /api/auth/status Verify authentication
POST /api/classify { "description": "..." } Classify AI risk level
POST /api/audit { "name": "...", "scores": [...] } Run 9-requirement audit
GET /api/scan?url=... Scan URL for compliance
POST /api/report { "audit_id": "..." } Generate report
POST /api/roadmap { "classification": "..." } Compliance roadmap
GET /api/stats Regulatory statistics
GET /api/search?q=... Search regulation
GET /api/knowledge Full regulatory database

⚖️ EU AI Act Quick Reference

Risk Levels

Level Color Examples Obligation
🔴 Prohibited Red Social scoring, subliminal manipulation Banned
🟠 High-Risk Orange CV screening, credit scoring, biometrics Full compliance (Art. 8-15)
🟡 Limited Yellow Chatbots, emotion recognition Transparency obligations
🟢 Minimal Green Spam filters, video games Voluntary codes

9 Mandatory Requirements (High-Risk AI)

# Requirement Article
R1 Risk Management System Art. 9
R2 Data Governance Art. 10
R3 Technical Documentation Art. 11
R4 Record-Keeping Art. 12
R5 Transparency Art. 13
R6 Human Oversight Art. 14
R7 Accuracy & Robustness Art. 15
R8 Cybersecurity Art. 15
R9 Quality Management Art. 17

Sanctions

Violation Fine
Prohibited AI €35M or 7% global revenue
High-risk non-compliance €15M or 3% global revenue
False information €7.5M or 1% global revenue

🤖 Platform Bots

# Telegram
export TELEGRAM_BOT_TOKEN="your-token"
python telegram_bot.py

# Slack
export SLACK_BOT_TOKEN="xoxb-your-token"
python slack_bot.py

# Discord
export DISCORD_BOT_TOKEN="your-token"
python discord_bot.py

# All at once
python start_all_bots.py

🛠 Tech Stack

Layer Technology
Backend Python 3.10+ (stdlib only — zero mandatory deps)
Frontend Vanilla HTML/CSS/JS, Material Design 3, PWA
Auth JWT (HMAC-SHA256), Google OAuth 2.0 (optional)
AI Engine DeepSeek V3 via OpenRouter (optional)
Deployment Docker, Nginx, Let's Encrypt
Bots Telegram, Slack, WhatsApp Cloud, Discord APIs

📅 Roadmap

  • Core classification engine (4 risk levels)
  • URL scanner with AI detection
  • 9-requirement audit (Articles 8-15)
  • Compliance report generator
  • Compliance roadmap builder
  • Knowledge base (complete regulation)
  • PWA with Material Design 3
  • JWT authentication (zero-config dev mode)
  • Multi-platform bots (Telegram, Slack, Discord, WhatsApp)
  • Marketing landing page
  • Automated demo video recording
  • Google OAuth production setup
  • PDF/DOCX report export
  • Multi-language support (EN/FR/DE/ES/IT)
  • Enterprise admin dashboard
  • Webhook notifications
  • CI/CD pipeline

🚀 Deployment

See deploy/README.md for Docker + VPS deployment.

# Quick Docker deploy
docker compose -f deploy/docker-compose.yml up -d

# With HTTPS
./deploy/enable_https.sh your-domain.com

👤 Author

Yacine Benhamou — AI Builder & EU Tech Compliance


📄 License

MIT License — See LICENSE for details.


Made with 🇪🇺 in Europe
The first open-source compliance toolkit for the world's first AI regulation

⭐ Star this repo · 🐛 Report Bug · 💡 Request Feature

About

120-Tool Autonomous AI Command Center - EU AI Act Compliance - Multi-Agent MCP + A2A Protocol - Nexus Control Center

Topics

Resources

Stars

Watchers

Forks

Packages

 
 
 

Contributors