Skip to content

mathisvfr/digital-assistant-quality-starter

Repository files navigation

Digital Assistant Quality Starter

A comprehensive starter kit for building high-quality digital assistants for the Dutch government (Nederlandse Overheid). This project provides datasets, best practices, and a full-stack platform to help developers create AI-powered assistants that comply with regulations and guidelines.

Word Cloud

Overview

This repository contains everything you need to develop a digital assistant that meets quality standards for public sector applications:

  • Curated Datasets: Extensive collection of Dutch government guidelines, regulations, and best practices
  • Best Practices: Documentation on AI governance, ethics, and compliance
  • Full-Stack Platform: Ready-to-use application with React frontend and FastAPI backend
  • Feedback System: Tools for collecting and incorporating user feedback

Project Structure

`

  1. Datasets/ # Dataset with scraped content and processed documents
  2. Best Practices/ # Best practices documentation
  3. Platform/ # Main application backend/ # Python FastAPI backend src/ # React frontend ...
  4. Feedback/ # Feedback documentation `

Quick Start

Prerequisites

  • Python 3.8+
  • Node.js 16+
  • GreenPT API Key

Backend Setup

`�ash cd "3. Platform/backend" cp .env.example .env

Add your GREENPT_API_KEY to .env

python -m venv venv

Windows:

.\venv\Scripts\activate

Linux/Mac:

source venv/bin/activate

pip install -r requirements.txt python start.py `

Frontend Setup

�ash cd "3. Platform" cp .env.example .env # Optional npm install npm run dev

Access the Application

Datasets

The 1. Datasets folder contains:

  • Scraped content from Dutch government sources
  • Processed markdown documents covering:
    • AI regulations and compliance (EU AI Act, AVG/GDPR)
    • Algorithm governance and risk management
    • Digital accessibility guidelines
    • Common Ground principles
    • Public sector IT standards (NORA, GEMMA)
    • And much more...

Features

  • RAG-based Responses: Leverages Retrieval-Augmented Generation for accurate answers
  • Source Citations: Provides references to official documentation
  • Compliance-Ready: Built with Dutch government regulations in mind
  • Modular Architecture: Easy to extend and customize

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Need Help?

Acknowledgments

  • Dutch Government (Rijksoverheid) for public documentation
  • VNG (Vereniging van Nederlandse Gemeenten) for Common Ground resources
  • All contributors to the open government ecosystem

About

Hackathon starter template - Gemeente AI Assistant met GreenPT

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors