How to take a full-page screenshot in Chrome, with or without an extension

Chrome can capture a whole page on its own; the command is just hidden in DevTools. Here is that method step by step, and when the print dialog, an extension or a script is the better tool.

You scrolled, you took a screenshot, you scrolled again, and now you have four images to glue together. Chrome has been able to capture an entire page for years. The command is simply in a place nobody looks. Here are four ways to do it, in order of how much you have to install, and a table at the end for choosing between them.

1. The built-in way: DevTools

No extension, no account. Chrome does this itself.

Open the page and scroll to the bottom once, so anything that loads on scroll has loaded.
Open DevTools: press F12, or Cmd+Option+I on a Mac and Ctrl+Shift+I on Windows and Linux.
Open the Command Menu: Cmd+Shift+P on a Mac, Ctrl+Shift+P on Windows and Linux.
Type screenshot and choose Capture full size screenshot.

A PNG of the whole page lands in your Downloads folder. The same menu has Capture screenshot for the visible part, Capture area screenshot for a region you drag, and Capture node screenshot for the element selected in the Elements panel. If you want the page at a particular width, say a phone's, toggle the device toolbar first, set the dimensions, and then run the same command.

What you get is a picture. It is exact, it is free, and you cannot select a word of it or search it. For a quick record of what a page looked like, that is often enough.

2. The print dialog: Save as PDF

Press Cmd+P or Ctrl+P, choose Save as PDF as the destination, and save. Under More settings Chrome offers the paper size, margins, scale, headers and footers, and whether to print background graphics.

This one keeps the text. You can search the PDF, copy from it and let a reader index it. The catch is that you get the page's print layout, not the page you were looking at. Sites ship a print stylesheet, and it may drop the navigation, change the column width, hide the comments or do nothing at all. Sometimes the print layout is exactly what you want. Sometimes it is a page you do not recognise, cut into paper pages wherever the ruler fell.

3. An extension: one click, and a choice of output

An extension is what you want when you do this more than once a week.

GoFullPage is the most used of them. It scrolls and stitches screenshots into one image, and exports PNG, JPG or a PDF with that image on a page. Capture and download are free; cropping, annotation, a date stamp and smart page splitting are in Premium, US$12 a year.
FireShot captures the whole page, the visible part or a selection, and its PDF can carry text and links; the text features, multipage PDFs and the editor are in Pro, US$39.95 a year or US$99.95 once.
Awesome Screenshot captures and also records the screen, shares by link and connects to Slack, Trello and the like. Its PDF goes through the print dialog.
Page Scanner lays the page out once at its full height and asks Chrome's print pipeline to render that layout, so the PDF's text is text and its artwork is vector, and the preview you crop is the same layout. Crop, annotate, blur or redact, choose A4, Letter or one long page, and extract. It is free. Chrome shows its debugging bar for the length of the scan, and it runs on Chrome and Chromium only.

A longer comparison of those four is here.

4. From a script, or from an AI agent

If the capture is part of a job rather than a moment, drive the browser instead of clicking it.

Puppeteer gives you page.screenshot({ fullPage: true }) and page.pdf(), which renders with the print stylesheet unless you emulate screen media first.
Playwright MCP exposes browser_take_screenshot with a full-page option and, as an opt-in, browser_pdf_save, to any agent that speaks MCP. By default it runs its own browser; its extension mode connects to the Chrome you are signed in to.
Chrome DevTools MCP exposes take_screenshot with a fullPage option in PNG, JPEG or WebP. It does not write PDFs.
Page Scanner has a command line, npx @page-scanner/cli scan --url … --out …, and an MCP server with a scan_page tool. Both run the capture in your own Chrome, through the extension, so a page behind a login is captured as you see it, and the PDF is the same vector PDF the editor makes.

The four of those are compared in a post of their own.

Which way, when

	You get	Text selectable	Looks like the page on screen	To install	Reach for it when
DevTools command	PNG	✗	✓	nothing	a one-off record of how a page looked
Print, Save as PDF	PDF on paper pages	✓	✗ the print layout	nothing	an article whose print layout is fine
GoFullPage	PNG, JPG, PDF of the image	✗	✓	an extension	frequent screenshots, any Chromium browser or Firefox
FireShot	PNG, JPG, PDF	✓ with Pro	✓	an extension	screenshots whose PDF should keep text and links
Awesome Screenshot	PNG, JPG, PDF via print	✗	✓	an extension	captures and recordings you share with a team
Page Scanner	PDF with text, PNG, JPEG	✓	✓	an extension	a document to search, print, annotate or redact
Puppeteer, the MCP servers	PNG or PDF	✓ for the PDFs	depends on the settings	code or an MCP client	captures inside a script or an agent's workflow

✓ yes, ✗ no.

Frequently asked questions

Why is the text in my screenshot fuzzy when I zoom in? Because it is a picture of text, at the resolution of your screen. If you need to zoom, search or copy, you need a PDF whose text is text: the print dialog, FireShot Pro or Page Scanner.

Why does the PDF from the print dialog look different from the page? The site's print stylesheet. Chrome prints what the site says a printed page should be, which can be narrower, plainer or missing parts.

Can I capture a page I am logged in to? With DevTools, the extensions and Page Scanner's command line and MCP server, yes: they all work inside your own Chrome. Puppeteer launches a browser of its own unless you connect it to a running one.

Is there a keyboard shortcut? GoFullPage and Page Scanner both answer to Alt+Shift+P, and Chrome lets you change any extension's shortcut at chrome://extensions/shortcuts.