How Cypress will revolutionize the use of AI in testing with cy.prompt()

Cypress has become the go-to testing framework for SDETs and QA engineers to validate modern web apps. It’s fast, reliable, and backed by a mature ecosystem—both in software updates and excellent documentation. Add to that the vibrant community building powerful plugins and extensions, and it’s clear why Cypress dominates the testing landscape.

Cypress is taking a bold step into AI-powered testing with the upcoming cy.prompt(). Unlike typical AI integrations that act as external copilots or rely on general-purpose MCP-style assistants, cy.prompt() adds the intent (what we want) built directly into the testing workflow.

This means no context switching, no juggling between an IDE plugin and your test runner. Instead, Cypress allows you to describe your intent in plain English, and the AI automatically generates selectors, actions, and assertions right inside your test.

It’s a shift from writing tests line by line to guiding your tests conversationally. Think less about cy.get() or cy.click() and more about telling Cypress what you want verified, letting the framework translate that into executable code.

Here’s a video demonstration of cy.prompt() in action:

This is the code that I used in the validation:

And here is what the prompt suggests of code locators right after is executed:

Also, you can leave the prompt as it is and push to your CI/CD pipeline:

With cy.prompt(), Cypress is no longer “just” a testing framework—it’s stepping into the AI-assisted development era. For SDETs and QA engineers, this means faster authoring, smarter locator handling, and easier onboarding for teams.

The possibilities of cy.prompt()

• Cy.prompt focus on the intent: what we want, not how do do it. It's a great tool for non-developers, or anyone who doesn’t want to dive deep into app implementation.
• Imagine writing the BDD (Behavior Driven Development) acceptance criteria directly into the test. You'll have the best of both worlds here, BDD criteria that is understood by all stakeholders (Project Managers, Product Owners), and the code being executed in the background.
• TDD (Test-Driven-Development) is also covered for the developers. Imagine developing a feature until is ready and, step by step (word by word, line by line) it start to pass. Until is ready for deployment.
• Portability is here: Need to refactor your project? Move from one programming language to another? Don't need to change a thing in your tests written in plain English, they can be easily shared, exported, or integrated across different systems.
• Also, another great benefit here are the self-healing tests, they’re more resilient to changes in the DOM or selectors. This feature could fundamentally change how we approach automation.

The future of QA is not just code—it’s collaboration between AI and testers.

What are your thoughts?

Share, comment or connect with me directly in LinkedIn!