> ## Content Index > Fetch the complete content index at: https://softwaretestingweekly.com/llms.txt > Use this file to discover other available public pages before exploring further. # Issue #305 - URL: https://softwaretestingweekly.com/issues/305/ - Published: 2026-03-11T07:00:00.000Z - Updated: 2026-06-10T13:19:21.000Z - Description: Great QA and Playwright Skills for AI Agents 🤖 - Author: Dawid Dylowicz | Comment | | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Welcome to the **305th** issue![Skills](https://support.claude.com/en/articles/12512176-what-are-skills?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) are the new way for instructing AI agents to generate better results.And now we have it for testing, too![QASkills.sh](https://qaskills.sh/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) by [Pramod Dutta](https://www.linkedin.com/in/pramoddutta/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)[Playwright Skills](https://skills.sh/currents-dev/playwright-best-practices-skill/playwright-best-practices?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) by [Walter Galvāo](https://www.linkedin.com/in/waltergalvao/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)[Playwright Skill Guides](https://github.com/testdino-hq/playwright-skill?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) by [Pratik Patel](https://www.linkedin.com/in/prat3ik/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)Happy (skills) testing! 🙂*[Dawid Dylowicz](https://linkedin.com/in/dawid-dylowicz/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)* | | | | NEWS | | [The Death of Determinism: How AI Forces Us to Rethink Testing](https://medium.com/capgemini-microsoft-team/the-death-of-determinism-how-ai-forces-us-to-rethink-testing-52bed49a8298?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)[Padget Avery](https://www.linkedin.com/in/padget-avery-80365a106/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) explains how testing AI differs from the traditional approach, requiring moving from strict assertions to thresholds, metrics, and human review in pipelines.Moreover, [Katja Obring](https://www.linkedin.com/in/katjaobring/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) shares thoughts on [What I don't understand about AI evals (yet)](https://kato-coaching.com/what-i-dont-understand-about-ai-evals-yet/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) and how [The AI evals field chose a flawed tool and stuck with it](https://kato-coaching.com/the-ai-evals-field-chose-a-flawed-tool-and-stuck-with-it/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[medium.com](https://medium.com/capgemini-microsoft-team/the-death-of-determinism-how-ai-forces-us-to-rethink-testing-52bed49a8298?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [We Need to Stop Queueing Quality](https://qualityeng.substack.com/p/stop-queueing-quality?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)Should teams keep a separate testing stage on their boards? [Jitesh Gosai](https://www.linkedin.com/in/jitgo/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) shares a simple analogy to show how it can delay feedback.Similarly, [Joshua Bihun](https://www.linkedin.com/in/joshuabihun/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) wrote good articles on [What Acceptance Criteria Actually Ought to Say](https://terriblefreedom.medium.com/what-acceptance-criteria-actually-ought-to-say-ed51ac224a52?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) and how to [Shift-Left Quality From Day One](https://terriblefreedom.medium.com/shift-left-quality-embedding-qa-thoughtfully-from-day-one-6314025e14f2?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[qualityeng.substack.com](https://qualityeng.substack.com/p/stop-queueing-quality?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [When building is cheap, quality becomes a bigger differentiator for your software product](https://medium.com/@bageresben/when-building-is-cheap-quality-becomes-a-bigger-differentiator-for-your-software-product-252d3fa0df7e?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)AI speeds up coding, but at what cost? [Esben Bager](https://www.linkedin.com/in/ebager/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) raises a good point — software advantage will now be driven primarily by quality and reliability rather than the number of features.Similarly, [Josh Grant](https://www.linkedin.com/in/josh-grant-9570a214/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) points out that [Software Quality is Heading Toward a Revolution](https://simplythetest.tumblr.com/post/809999894733619200/software-quality-is-heading-toward-a-revolution?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) and, in a similar context, [Maaike Brinkhof](https://www.linkedin.com/in/maaike-brinkhof-1942b725/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) says to [Stop selling gold when all you've got to offer is a process](https://www.maaikebrinkhof.nl/stop-selling-gold-when-all-youve-got-to-offer-is-a-process/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[medium.com](https://medium.com/@bageresben/when-building-is-cheap-quality-becomes-a-bigger-differentiator-for-your-software-product-252d3fa0df7e?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [Why I am not worried about AI replacing me](https://medium.com/@tldr%5Fhow%5Fto%5Ffix%5FQA/why-i-am-not-worried-about-ai-replacing-me-d9a7b44094a0?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)With the rapid, constant development of AI, you may wonder whether it poses a threat to the tester's role. [Matthew Sullivan](https://www.linkedin.com/in/matthew-sullivan-309b0722/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) doesn't think so, for a few reasons.Similarly, [Brad DerManouelian](https://www.linkedin.com/in/brad-dermanouelian/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) explains why [AI Isn't Killing Manual QA. It's Supercharging It](https://docs.testplanit.com/blog/ai-isnt-killing-manual-qa/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[medium.com](https://medium.com/@tldr%5Fhow%5Fto%5Ffix%5FQA/why-i-am-not-worried-about-ai-replacing-me-d9a7b44094a0?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | | | AUTOMATION | | [AI and Testing: Using Local Models for Testing](https://testerstories.com/2026/02/ai-and-testing-using-local-models-for-testing/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)[Jeff Nyman](https://www.linkedin.com/in/jeffnyman/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) continues a detailed series of articles on AI in testing, and shows how to set up and run a model on your machine to create automated tests with evaluation metrics. You can also read a follow-up article about [Using Model Pipelines for Testing](https://testerstories.com/2026/03/ai-and-testing-using-model-pipelines-for-testing/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[testerstories.com](https://testerstories.com/2026/02/ai-and-testing-using-local-models-for-testing/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [My Engineering Manager asked for a test coverage dashboard. Here's what I built](https://medium.com/automationmaster/my-engineering-manager-asked-for-a-test-coverage-dashboard-heres-what-i-built-55b81b789327?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)Who wasn't in that situation? It's interesting to see how [Osanda Deshan Nimalarathna](https://www.linkedin.com/in/osandadeshan/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) approached building a dashboard for tracking automated test coverage.[medium.com](https://medium.com/automationmaster/my-engineering-manager-asked-for-a-test-coverage-dashboard-heres-what-i-built-55b81b789327?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [Only 41% of Testing Is Automated in 2026\. More AI Tools Won't Fix That.](https://scrolltest.medium.com/only-41-of-testing-is-automated-in-2026-more-ai-tools-wont-fix-that-daebca82e7dd?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)[Pramod Dutta](https://www.linkedin.com/in/pramoddutta/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) shares insights from the recent Leapwork survey on the state of test automation, revealing some (un)surprising data.[scrolltest.medium.com](https://scrolltest.medium.com/only-41-of-testing-is-automated-in-2026-more-ai-tools-wont-fix-that-daebca82e7dd?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [Rethinking performance testing for the modern web](https://medium.com/refluent/rethinking-performance-testing-for-the-modern-web-with-browsers-338d0b84cf74?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)An insightful article from [Shyamal Raju](https://www.linkedin.com/in/shyamalraju/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) explaining why browser performance is important for user experience and demonstrating how to include browser flows in load testing.On this note, [Oleh Koren](https://www.linkedin.com/in/oleh-koren/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) shares good thoughts on [Why Percentiles Matter More Than Average Response Time in Performance Testing](https://medium.com/@oleh.koren96/why-percentiles-matter-more-than-average-response-time-in-performance-testing-529c9e235d8f?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[medium.com](https://medium.com/refluent/rethinking-performance-testing-for-the-modern-web-with-browsers-338d0b84cf74?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [Test Plan Driven Development (TPDD): A Developer-Friendly Alternative to TDD](https://medium.com/@vigorzjw/test-plan-driven-development-tpdd-a-developer-friendly-alternative-to-tdd-02f393048ff6?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)This sounds novel. [Vigor Zang](https://www.linkedin.com/in/vigorz/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) recognises the friction with practising Test-Driven Development and suggests a different approach to ensure quality since the early stages of the software development life cycle.Moreover, [Sefa Esendemir](https://www.linkedin.com/in/sefaesendemir/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) wrote about more levels of the test pyramid — [From Writing Code to Breaking It](https://medium.com/@sefaesendemir/from-writing-code-to-breaking-it-0276c3be75df?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[medium.com](https://medium.com/@vigorzjw/test-plan-driven-development-tpdd-a-developer-friendly-alternative-to-tdd-02f393048ff6?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | | | TOOLS | | [17 Playwright Testing Mistakes You Should Avoid](https://elaichenkov.github.io/posts/17-playwright-testing-mistakes-you-should-avoid/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)There is some great, practical advice on how to improve your [Playwright](https://playwright.dev/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) tests to run more smoothly and fail less often. By [Yevhen Laichenkov](https://www.linkedin.com/in/elaichenkov/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[elaichenkov.github.io](https://elaichenkov.github.io/posts/17-playwright-testing-mistakes-you-should-avoid/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [Deep Dive into Playwright CLI: Token Efficient Browser Automation](https://testdino.com/blog/playwright-cli/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)Two weeks ago, I [mentioned](https://softwaretestingweekly.com/issues/303?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305#start) the release of [Playwright CLI](https://github.com/microsoft/playwright-cli?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305). Now we can learn more about it, thanks to this detailed overview by [Pratik Patel](https://www.linkedin.com/in/prat3ik/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[testdino.com](https://testdino.com/blog/playwright-cli/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [k6-cucumber-steps: Run K6 Performance Tests with Cucumber BDD Syntax](https://medium.com/@paschal.enyimiri%5F34659/k6-cucumber-steps-run-k6-performance-tests-with-cucumber-bdd-syntax-c5bdca09bae9?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)Cucumber for performance tests? [Chetachi Enyimiri](https://www.linkedin.com/in/chetachi-enyimiri-05237a144/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) thought "Why not!" and demonstrated how to achieve that with [k6](https://k6.io/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).Also, [Marlo Henrique](https://www.linkedin.com/in/marlo-henrique-84a940109/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) advises [Using SharedArray with k6](https://medium.com/@marloh2222/using-sharedarray-with-k6-233876dd3b8b?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[medium.com](https://medium.com/@paschal.enyimiri%5F34659/k6-cucumber-steps-run-k6-performance-tests-with-cucumber-bdd-syntax-c5bdca09bae9?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [Making Code Coverage Work with Playwright](https://medium.com/@afnank/making-code-coverage-work-with-playwright-for-non-vite-setups-a1028866b300?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)[Sayed Afnan Khazi](https://www.linkedin.com/in/sayed-afnan-khazi/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) tells us how they've been looking for a way to get Playwright test coverage in a non-trivial project setup and how the [playwright-coverage](https://github.com/bgotink/playwright-coverage?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) library made it happen.Speaking of that, [Bhavik Gandhi](https://www.linkedin.com/in/gandhi-bhavik/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) wrote a helpful guide on [How to Intercept and Modify Network Responses with Playwright](https://qaclub.io/bhavik/how-to-intercept-and-modify-network-responses-with-playwright/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[medium.com](https://medium.com/@afnank/making-code-coverage-work-with-playwright-for-non-vite-setups-a1028866b300?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | [Migrating From Cypress.env To cy.env and Cypress.expose Methods](https://glebbahmutov.com/blog/cypress-expose/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)There's a major change coming in [Cypress](https://www.cypress.io/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) v16 and [Gleb Bahmutov](https://www.linkedin.com/in/bahmutov/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) shows how to prepare for handling the environment variables and secrets in a new way.Also, [Jennifer Shehane](https://www.linkedin.com/in/jennifer-shehane/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) has just announced another improvement that allows you to [Fix Cypress CI failures caused by no spec files found](https://www.cypress.io/blog/fix-cypress-ci-failures-caused-by-no-spec-files-found?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305).[glebbahmutov.com](https://glebbahmutov.com/blog/cypress-expose/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) | | | | And… | | ["Quality is too high..."](https://www.linkedin.com/feed/update/urn:li:activity:7437066151509102592/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305) 🤣 | | | | Thanks for reading!If you enjoy this newsletter and find it helpful in becoming a better tester, please consider sharing it with others.*[Dawid Dylowicz](https://linkedin.com/in/dawid-dylowicz/?utm%5Fcampaign=Software%2BTesting%2BWeekly&utm%5Fsource=Software%5FTesting%5FWeekly%5F305)* |