Skip to content

Yacinewhatchandcode/EU-AI-Act-Compliance

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

12 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

Prime-AI Logo

πŸ‡ͺπŸ‡Ί Prime-AI β€” 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 β€” Landing Page β€’ Auto-Login β€’ Risk Classifier β€’ URL Scanner β€’ 9-Requirement Audit β€’ Knowledge Base


πŸ’‘ What is this?

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

Feature Description
πŸ” URL Scanner Scan any website to detect AI systems and assess compliance risk
βš–οΈ Risk Classifier Classify AI systems into 4 levels: Prohibited β†’ High β†’ Limited β†’ Minimal
πŸ“‹ 9-Requirement Audit Full audit against Articles 8-15 with weighted scoring
πŸ“Š Compliance Reports Generate reports with remediation roadmaps
πŸ“š Knowledge Base Complete regulatory database β€” 8 prohibited, 8 high-risk, 9 requirements
πŸ€– 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 β€” 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 β†’ http://localhost:8080

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


🎯 Use Cases β€” 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.


πŸ—οΈ Architecture

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚              Client Layer (PWA / Bots)              β”‚
β”‚   Web App  Β·  Telegram  Β·  Slack  Β·  WhatsApp  Β·  Discord β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                       β”‚ HTTP / WebSocket
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”΄β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚              eu_ai_act_server.py (API)              β”‚
β”‚   JWT Auth  Β·  REST API  Β·  Static Files            β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                       β”‚
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”΄β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚              eu_ai_act.py (Core Engine)             β”‚
β”‚   Classifier  Β·  Auditor  Β·  Scanner  Β·  Reporter  β”‚
β”‚   Regulatory DB  Β·  Risk Matrix  Β·  Remediation     β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

Project Structure

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

πŸ”Œ API Reference

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

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

βš–οΈ EU AI Act Quick Reference

Risk Levels

Level Color Examples Obligation
πŸ”΄ 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

πŸ€– 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

πŸ›  Tech Stack

Layer Technology
Backend Python 3.10+ (stdlib only β€” 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

πŸ“… 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

πŸš€ 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

πŸ‘€ Author

Yacine Benhamou β€” AI Builder & EU Tech Compliance


πŸ“„ License

MIT License β€” See LICENSE for details.


Made with πŸ‡ͺπŸ‡Ί in Europe
The first open-source compliance toolkit for the world's first AI regulation

⭐ Star this repo Β· πŸ› Report Bug Β· πŸ’‘ Request Feature

About

πŸ‡ͺπŸ‡Ί Full-Stack EU AI Act Compliance Platform β€” Risk Classification, Audit Engine, Knowledge Base | Python Β· Flask Β· PWA

Topics

Resources

Stars

Watchers

Forks

Packages

 
 
 

Contributors