Tired of PowerShell scripts that can’t interact with a webpage? Learn three methods, from the simple `Start-Process` to powerful automation with Selenium and Playwright, to finally control a web browser from your scripts.

So You Want to Drive a Web Browser with PowerShell? Let’s Talk.

I remember it like it was yesterday. It was 2 AM, and a P1 alert was screaming about high latency on our legacy order processing system. The only way to get the *real* performance metrics was to log into this ancient, crusty Java-based web portal on `legacy-monitor-01`, click three specific buttons, and look at a graph that looked like it was designed in 1998. There was no API. No export function. Just a screen. My brilliant idea? “I’ll just whip up a PowerShell script to open the page, log in, and take a screenshot every 5 minutes!” An hour later, I was pulling my hair out, staring at `Invoke-WebRequest` results that were just a mess of useless login page HTML, wondering why something so simple felt so impossible. If that sounds familiar, you’re in the right place.

First Off, Why Is This So Hard?

Let’s get one thing straight: PowerShell is a shell, not a web browser. When you use commands like Invoke-WebRequest or Invoke-RestMethod, you’re just sending a raw HTTP request and getting back the raw HTML, JSON, or whatever the server sends. You aren’t running any of the JavaScript, you aren’t rendering any CSS, and you certainly can’t “click” a button. To do that, you need to remote-control an actual browser. This requires a middleman—a “WebDriver”—that acts as a translator between your script’s commands (like “Go to this URL” or “Find the element with ID ‘login-btn’”) and the browser’s internal automation engine.

The Solutions: From Simple to Powerful

We’ve all been there, so let’s break down the options from the quick-and-dirty fix to the “let’s build a real automation pipeline” solution.

Solution 1: The Quick & Dirty “Fire and Forget”

Sometimes you don’t need to control the page; you just need to open it. Maybe you’re writing a script for a junior admin and you want to pop open the Grafana dashboard for them at the end. For that, the built-in Start-Process cmdlet is your best friend. It does exactly what it says: it starts a process. No control, no feedback, just pure, simple execution.

# Just opens the URL in your default browser. That's it.
Start-Process -FilePath "https://monitoring.techresolve.com/dashboards/db/production-overview"

# You can even specify the browser if you know the executable path
Start-Process -FilePath "C:\Program Files\Google\Chrome\Application\chrome.exe" -ArgumentList "https://google.com"

Pro Tip: This is a one-way street. Your script launches the browser and then moves on. It has no idea what happens in that browser window after it opens. Don’t use this if you need to wait for a result or interact with the page.

Solution 2: The “Industry Standard” Way with Selenium

This is the answer for 90% of real-world browser automation tasks. Selenium has been the bedrock of web automation and testing for years. It’s reliable, powerful, and has a massive community. The concept is simple: you install a PowerShell module for Selenium and a separate WebDriver executable that matches your browser and version.

Here’s how you get it done:

1. Install the PowerShell Module: Install-Module -Name Selenium -Repository PSGallery -Force
2. Download the WebDriver: You need the driver for your specific browser. For Chrome, it’s ChromeDriver. For Edge, it’s EdgeDriver. IMPORTANT: The driver version MUST match your installed browser version closely!
3. Write the Script:

# Point to the location of your downloaded chromedriver.exe
$driverPath = "C:\Tools\WebDrivers\"
$driver = New-Object OpenQA.Selenium.Chrome.ChromeDriver($driverPath)

# Navigate to the site
$driver.Navigate().GoToURL("https://portal.techresolve.com/login")

# Find elements and interact with them
$usernameField = $driver.FindElementById("username")
$usernameField.SendKeys("darian.vance")

$passwordField = $driver.FindElementById("password")
$passwordField.SendKeys("MySuperSecretPassword123") # Obviously, use secure methods for this!

$loginButton = $driver.FindElementByTagName("button")
$loginButton.Click()

# Wait a bit for the next page to load (better methods exist, but this is simple)
Start-Sleep -Seconds 5

# Get the title of the new page to confirm login
Write-Host "Landed on page with title: $($driver.Title)"

# Always clean up!
$driver.Quit()

Warning: The biggest headache with Selenium is keeping the WebDriver executable in sync with your browser. Chrome updates automatically and silently, and your script will suddenly break with a cryptic error. You’ll need a process to periodically check and update your drivers.

Solution 3: The “Modern & Robust” Way with Playwright

Playwright is the new kid on the block from Microsoft, and frankly, it’s fantastic. It was built from the ground up for modern web apps (think React, Vue, etc.) and it solves many of Selenium’s classic pain points. The best part? It manages the browser and driver installations for you!

It takes a bit more setup initially as it’s built on Node.js, but the PowerShell integration is excellent.

1. Install the PowerShell Module: Install-Module -Name Playwright -Repository PSGallery -Force
2. Run the Install Command: This is the magic. It downloads the browsers and drivers for you. Install-Playwright
3. Write the Script: The syntax is a bit different, but it’s clean and modern.

# Import the module
Import-Module Playwright

# This starts Playwright and opens a browser
$p = Start-Playwright
$browser = $p | New-Browser -Browser 'chromium' -Headless:$false # Set Headless to $true to run in the background
$page = $browser | New-Page

# Navigate and interact
$page | Go-To-Page "https://portal.techresolve.com/login"
$page | Fill-Input '#username' 'darian.vance'
$page | Fill-Input '#password' 'MySuperSecretPassword123'
$page | Click-Element 'button'

# Playwright has built-in "waits" that are smarter than Start-Sleep
$page | Wait-For-Navigation

# Take a screenshot to verify
$page | Take-Screenshot -Path 'c:\temp\dashboard.png' -FullPage

# Clean up
$p | Stop-Playwright

Which One Should You Choose? A Quick Comparison

My Final Two Cents

Look, if you just need to pop a URL open, use Start-Process and call it a day. For anything more, I strongly recommend putting in the effort to learn Playwright. I’ve been migrating our older Selenium scripts over, and the stability improvements and reduced maintenance (no more manual driver updates!) have been a lifesaver. Selenium is still a great and valid tool, but Playwright feels like it was designed for the kinds of problems we DevOps engineers face every day. Don’t let the initial setup scare you off; it’ll pay for itself the first time you don’t get a 3 AM page because Chrome updated and broke your script.

Darian Vance

Lead Cloud Architect & DevOps Strategist

With over 12 years in system architecture and automation, Darian specializes in simplifying complex cloud infrastructures. An advocate for open-source solutions, he founded TechResolve to provide engineers with actionable, battle-tested troubleshooting guides and robust software alternatives.

🤖 Frequently Asked Questions