Validating behavior after AI changes your code

UI testing for Power Platform has always been painful - lack of tooling, a lot of upfront effort, constant maintenance and tests that break with every form change. This session shows a different approach: using LLMs to generate stable, maintainable test code from natural language scenarios.

Why UI-level tests over unit tests? When AI writes code, it writes tests that pass - adjusting both together until you get green checkmarks without real validation. User behavior simulation is the honest feedback loop: tests describe what users expect, not what the code does.

The workflow: an LLM with Playwright MCP explores your app, identifies testable scenarios, and generates Gherkin test plans. A GitHub Copilot agent converts those into Playwright code that runs without AI in your pipeline - deterministic, no inference costs. When tests fail, agent mode analyzes failures and proposes fixes.

Gherkin scenarios are implementation-agnostic - they describe behavior, not selectors. This makes them stable when UI changes, and lets them serve dual purposes: executable specifications that guide AI coding agents, and test definitions that validate the result. For new apps, you generate scenarios from specifications before any code exists - test-driven development with AI-generated tests as quality gates.

Oficiální Power Platform Skills a jak je předat agentovi

Ukážeme si jak můžeme vytvořit aplikaci pomocí oficiálních power platform skills, jak nainstalovat plugin a jak to vše funguje dohromady

Infrastructure as Code v Power Platform pomocí Terraform

Ukážeme si co je to IaC, jak se používá nástroj Terraform a jak to všechno můžeme používat v Power Platformě

Automatické generování testů přes playwright cli

Ukážeme si jak můžeme generovat testy pro code apps pomocí playwright cli, abychom měli jistotu že nám copilot nerozbíjí kod pod rukama

Power FX SDK Lessons Learned

UDPP 2025 - Developer Lab - Source Control, Build System and CI/CD

Learn how to improve developer productivity through continuous integration, automation and testing frameworks. This workshop focuses on building sustainable, efficient workflows that support long-term success in Power Platform development.

https://power2025.updatedays.cz/workshop/en

by Matěj Samler (together with Scott Durow, Eric Regnier & Parvez Ghumra)

EPPC 2025 - Source Control, Build System and CI/CD Pipelines

https://www.sharepointeurope.com/events/source-control-build-system-and-ci-cd-pipelines/

PCF in 15 minutes

Ukážeme si, jak na PCFs a jedno si vytvoříme v přímém přenosu.

Power Cards a jak je použít

Ukážeme si příklad, jak použít novou komponentu PP: Power Cards. Jaké jsou její silné i slabé stránky a proč je zvážit oproti adaptive cards

GitHub Actions for Power Platform

Ukážeme si, jak automatizovat CI/CD nad Power Platform pomocí Github Actions

EPPC 2025 - Developer ALM Lab – Source Control, Build System and CI/CD

https://www.sharepointeurope.com/events/developer-alm-lab-source-control-build-system-and-ci-cd/

UDPP2025 - Integrations with Custom Connectors, Dataflows and Virtual Tables

- Building and using Custom Connectors to connect to external APIs and services
- Using Dataflows for low-code ETL to ingest, transform and map data into Dataverse
- Virtualizing external data with Virtual Tables to enable access without duplication
- Comparing use cases, limitations and best practices
- Enabling full CRUD operations with custom virtual table data providers

https://power2025.updatedays.cz/schedule/en/integrations-with-custom-connectors-dataflows-and-virtual-tables

-with Jan Koštejn

UDPP 2026 - Setting up developer environment, repository and build system

Focus on the developer inner loop - the fast local cycle before code reaches CI/CD. Covers source controlling Dataverse solutions, setting up MSBuild-based local builds, generating version numbers from Git history, working with early bound types, running validation locally, unit testing plugins and PCF controls, and reducing dependency on long-lived development environments.

https://power.updatedays.cz/schedule/en/setting-up-developer-enviroment-repository-and-build-system

by Matěj Samler and Tomáš Prokop

UDPP2026 - Validating behavior after AI changes your code

Two complementary testing approaches for Power Platform: TDD with mocking frameworks for frontend and backend code, plus AI-assisted BDD acceptance tests that explore applications via browser automation. With Code Apps, Generative Pages, and vibe coding, UIs will change more frequently than ever - natural language scenarios provide stable behavioral contracts when UI abstractions keep shifting.

https://power.updatedays.cz/schedule/en/validating-behavior-after-ai-changes-your-code

by Matěj Samler and Julie Koťátková