Selenium MCP server

This is a helper that lets your AI open a real browser, click around a website, and check that things look right. It is handy if you test web pages and want the boring parts written for you. You talk to your AI in plain words, and it does the clicking for you.

What this MCP server does

You tell your AI something like go to this page and log in. Your AI sends that to this helper, and the helper opens Chrome and does the steps. Every action is recorded along the way. When you ask, the helper turns that recording into test code in Java, Python, or C#. You can also ask it to check that a heading is visible or that a button is clickable, and it tells you yes or no.

Click to zoom

What you can do with it

• Open a website and click through it like a real person
• Fill in forms, pick from dropdowns, and upload files
• Check that text, buttons, and headings are where they should be
• Take screenshots and compare them to an earlier one
• Record your steps and turn them into test code
• Generate CI pipeline files for GitHub, Jenkins, or GitLab
• Run a basic accessibility check on a page

Try asking your AI

• “Go to https://myapp.com and fill the login form with admin and password, then click Login”
• “Assert the dashboard heading is visible”
• “Generate a Java TestNG test class for everything we just did”
• “Generate a Gherkin feature file and step definitions for the login flow”

What it gives back to you

You get answers in the chat: page titles, URLs, text from elements, and pass or fail results for your checks. Screenshots come back as images you can see right in the conversation. When you ask for code, you get a full file you can copy into your project. It also keeps a log of everything it did, so you can look back at the steps.

Before you start

Who it's for

Testers, QA engineers, and developers who check web pages and want help writing or fixing automated tests.