Issue 287

COMMENT

Welcome to the 287th issue!

Playwright has just announced the introduction of Agents!

The three Planner, Generator and Healer test agents are responsible for guiding AI through the core process of building a Playwright test.

Sounds exciting, doesn't it?!

Karthik K.K. has already tried them out and recorded a 30-minute demo, followed by an explanation that while the solution has potential, it may not be ready yet for some use cases.

It's an intriguing move nevertheless and we'll likely see more development and improvements in this area.

Happy testing! 🙂

NEWS

Every Bug Deserves a Test Case

How do you prevent bugs from happening again? Kevin Konda suggests a discipline of covering past bugs with test cases, and it's not a bad idea.

At the same time, Do You Really Need "Classic" Detailed Test Cases Today?

Seriously Testing LLMs

What does it take to test LLMs? After a series of experiments, James Bach came up with a LARC method and explained why testing thoroughly is challenging and may get expensive.

The Day I Became an AI "Babysitter" (And Why I'm Not Ashamed of It)

This is a great deep dive into using AI in testing by Santhosh Siddegowda, based on the lessons learned from transforming traditional QA test cases into AI-assisted ones.

Web Accessibility Testing — Are We There Yet?

With the European Accessibility Act going into effect recently, testing for accessibility has become more important than ever. But Bhagya Perera points out that there's still room for improvement in this regard.

On top of that, Chris Yoong raises a good point — Why does big tech not run Accessibility bug bounties?

"Why didn't testing find this issue?" Because you desire something that doesn't exist!

Ah, that question when things go wrong! Maaike Brinkhof shares the right response to it.

Also, Jeff Nyman makes a rather unusual but intriguing point that Testing Has Something To Do With Mass Extinction. Or in other words — is testing dying?

AUTOMATION

API Testing vs Browser Automation

Ever wondered whether you should automate API or UI tests for web apps? Someone shared their thoughts about the hybrid approach with the testing community.

And once you have the tests, you may want to Organise Your Tests in CI/CD as per Mariam Sargsyan's advice.

Did we automate all the tests yet? The manual testing reality few people talk about

Cuneyt Bati raises a good point about maintaining a balance between automated and non-automated types of tests, despite some environments' conviction towards full automation.

Secrets Behind 3 Years of Automation Success

Nikolay Advolodkin invited Oles Nikaniuk to share about their long-term test automation strategy, covering the choice of tools, types of testing, CI/CD setup, and much more.

Test Automation: How to Turn Regression Routine into a Reliable System

Maksim Laptev shares a detailed story about moving from manual testing to automation, including advice on prioritising test cases, selecting tools and building a scalable test architecture.

Furthermore, Vitaly Sharovatov tells us why we should treat Test Automation as Investment, Not Just Technical Choice.

The Day I Stopped Trusting My Load Tests (And Started Simulating Chaos Instead)

Hariprasath V S describes how applying the Monte Carlo method helped them expose the risks of unpredictable user behaviour breaking the system under load.

TOOLS

Cypress vs Playwright; Browser Included

Do you know how exactly Cypress and Playwright interact with the browser? Gleb Bahmutov breaks it down in detail with examples.

How Playwright Runs Workers and Test Fixtures (Parallel vs Serial vs Default)!

Thananjayan Rajasekaran continues sharing insightful articles about Playwright, this time covering the different options of running tests in parallel with helpful code snippets.

What's new in JUnit 6: Key Changes and Improvements

JUnit 6 is here! It's the first major release in years, and Vladimir Dmitrienko tells us more about what's new.

Writing custom Cypress plug-ins that solve common software testing problems

Cypress is known for its extensive plugin library, which anyone can contribute to. Want to know how? Kanika Vatsyayan demonstrates it with examples.

VIDEOS

Chrome DevTools MCP: Automated Test, Debug and Performance Analysis with AI

Google has recently released the Chrome DevTools MCP and Karthik K.K. recorded a handy 18-minute overview of how it enables AI agents to run and debug tests.

How should I test my AI prompts?

Apart from testing LLM models, you may also want to test your prompts. Daniel Knott gives a good overview of how to approach that in this 20-minute video.

Moreover, Jatin Parate shares more about The Ultimate Prompt: How I Get Perfect Backend Tests from AI in One Shot.

AND...

Load testing the finance department... 😅

Thanks for reading!

If you enjoy this newsletter and find it helpful in becoming a better tester, please consider sharing it with others.

Issue 287 October 13th 2025