Devlog 2026-06-07 · 7 min read

How Our AI Agents Built a ZUGFeRD & E-Invoicing XML Generator in Record Time

Explore the technical journey of how our AI agents Jan and Klára designed and programmed a modern utility for generating ZUGFeRD and XRechnung hybrid electronic invoices, fully compliant with the European EN 16931 standard.

Devlog: ZUGFeRD & E-Invoicing XML Generator

In a world of rapid digital advancement, staying compliant with regulatory requirements, especially in financial transactions, is key. We are proud to introduce our latest creation – the ZUGFeRD & E-Invoicing XML Generator. This tool is designed to simplify and automate the process of creating valid hybrid electronic invoices that conform to the stringent requirements of the European EN 16931 e-invoicing standard, including ZUGFeRD and XRechnung formats.

The Technical Challenge: The Complexity of E-Invoicing

Generating electronic invoices that are both machine-readable (XML) and human-readable (PDF) is a complex task. The EN 16931 standard defines a comprehensive set of rules for data structure, formatting, and semantics. Manually creating such invoices is time-consuming and prone to errors. Our goal was to build an intuitive tool that abstracts this complexity, allowing developers and businesses to easily generate compliant e-invoices.

AI Agents in Action: Jan, Klára, Martin, and Tomáš

Our internal team of AI agents contributed to the development of this generator, each with their specific role:

  • Jan (AI Developer): Jan was responsible for the architectural design and implementation of the generator's core. He focused on parsing user inputs, dynamically generating the XML structure according to the EN 16931 schema, and robust multilingual support. His work ensured that the generated XML is syntactically and semantically correct, and that the widget is globally usable.

> "A key element for global usability was the robust implementation of i18n," explains Jan, our AI Developer. "We had to ensure that all labels and messages in the interface would be dynamically localized, allowing users from different countries to feel comfortable and effectively create invoices in accordance with local regulations and language preferences. This involved careful management of translation keys and their mapping to individual language environments, as seen in the introduction of our i18n object:

```javascript

const i18n = {

en: {

widgetTitle: "ZUGFeRD & E-Invoicing XML Generator",

invoiceMetadata: "Invoice Metadata",

invoiceId: "Invoice ID:",

invoiceIdPlaceholder: "e.g., INV-2023-001",

invoiceDate: "Invoice Date:",

dueDate: "Due Date:",

currency: "Currency:",

senderDetails: "Sender Details",

senderName: "Sender Name:",

senderNamePlaceholder: "e.g., Your Company Ltd.",

senderAddress: "Sender Address:",

senderAddressPlaceholder: "e.g., Street 1, City",

senderVAT: "Sender VAT ID:",

senderVATPlaceholder: "e.g., GB123456789",

recipientDetails: "Recipient Details",

recipientName: "Recipient Name:",

recipientNamePlaceholder: "e.g., Client Corp.",

recipientAddress: "Recipient Address:",

// ... and more multilingual translations

```

  • Klára (AI Designer): Klára designed the user interface (UI) and user experience (UX). She focused on a clean, intuitive design that allows users to easily input invoice metadata, tax settings, and line items. Her work also included the visualization of a live XML preview and a styled PDF mockup for immediate feedback.
  • Martin (AI QA Engineer): Martin handled thorough functional testing. He validated the generated XML against official schemas, verified calculation correctness, and tested integration with the PDF generator. His meticulousness ensured that the tool produces error-free and standard-compliant results.
  • Tomáš (AI DevOps Engineer): Tomáš ensured the seamless deployment of the generator. He set up the CI/CD pipeline, optimized performance, and monitored operations to guarantee high availability and reliability.

Key Features of the Generator

Our ZUGFeRD & E-Invoicing XML Generator offers a robust set of features:

  • Intuitive Metadata Input: Easy entry of invoice ID, issue date, due date, and currency.
  • Comprehensive Sender and Recipient Details: Detailed information about invoicing parties, including addresses and VAT IDs.
  • Configurable Tax Settings: Flexible management of VAT rates and their application to individual line items.
  • Invoice Line Item Management: Add, edit, and delete line items with detailed descriptions, quantities, unit prices, and discounts.
  • Live XML Schema Preview: Instant visualization of the generated XML, allowing developers to monitor data structure in real-time.
  • Styled PDF Mockup Preview: Immediate visual representation of the invoice in a readable format.
  • WhatsApp Integration for Download: A unique feature for instantly downloading the compiled hybrid PDF/XML e-invoice directly to WhatsApp via our accounting assistant, Eva.

Try It Out for Yourself!

We are thrilled to share this tool with the community. Whether you are a developer seeking a robust e-invoicing solution or a business owner needing to streamline your administration, our generator will help you. Try the live demo and see for yourself its power and simplicity:

Live Demo: https://pixeloffice.eu/showcase/demo-webtrh-zugferd-xml-invoice-generator/

Let us know what you think! Your feedback is invaluable to us as we continue to improve our AI tools.

🏢

Pixel Office

Explore the ZUGFeRD & E-Invoicing Generator and streamline your workflow!

💬 WhatsApp

More articles

AI Development

Devlog: AI Voice Agent Playground – A Revolution in B2B Communication

At Pixel Office, we proudly launched our latest lead magnet, the "AI Voice Agent Playground," allowing businesses to experience the power of AI voice assistants. Dive with us into the technical details and team collaboration behind this innovative tool.

Read more →
Devlog

New Devlog: WhatsApp CMS Gallery – A Revolution in Real-Time Project Updates

Explore the technical details of our innovative lead magnet, allowing businesses and craftsmen to update photo galleries on their website in real-time directly from WhatsApp. Discover how Socket.IO, Gemini Vision, and our AI teams collaboratively built a tool that transforms how you share your work.

Read more →
Devlog

How Our AI Agents Built a Robust UTM Campaign Link Builder in Record Time

Dive into the technical details of how our AI agents Jan and Klára collaborated to create an interactive UTM Campaign Link Builder with live preview and QR code generation, and how Martin and Tomáš brought it to perfection.

Read more →