pagevera

PageVera: save any web page as a clean PDF

PageVera is a Chrome extension that saves any web page as a clean, high-quality PDF in one click. The name joins “page” with “vera”, Latin for true, because the PDF stays true to the original page. Navigation bars, sidebars, cookie banners and chat bubbles are left out. The text stays real text you can search and copy, and the images keep their original resolution.

The problem

Chrome can already save a page as a PDF, but only through the print dialog, and it prints the page exactly as it is at that moment. On many sites that means:

• Headers, sidebars, cookie banners and chat bubbles print on top of the content or repeat on every page.
• Images that only load as you scroll come out blank or blurry.
• Carousels and slideshows print only the slide that happens to be showing.
• Images and slides get cut in half at page breaks, and some PDFs start with a blank page.
• Lessons and documents shown inside embedded players are cut off at the edge of the player’s window.

Who it’s for

People who keep web content to read later, annotate or archive: students saving course lessons, analysts and researchers saving articles and documentation, and anyone who reads offline. I built it for my own use while taking an online course, to keep a PDF of each lesson next to its video.

What it does

• Saves with one click from the toolbar, a keyboard shortcut or the right-click menu, with no print dialog.
• Removes the top navigation, sidebars, footer, popups, cookie banners and chat widgets, then lets the content use the freed space.
• Lets you click any element on the page to leave it out, with a preview of what the automatic removal will take.
• Scrolls to the end first, so images that load on scroll and infinite feeds are included.
• Keeps text selectable and searchable, keeps links clickable, and embeds images at their original resolution.
• Prints every slide of a carousel, slideshow or image gallery one after another, without the arrows and dots.
• Uses normal paper pages (Letter or A4 depending on your region) and keeps images and slides whole at page breaks. One long page is also available.
• Works inside embedded course players and document viewers, including ones loaded from other sites.
• Names the file after the page, or after the lesson inside a course player. If Chrome asks where to save, its dialog opens in the last folder you used.

Product decisions

Chrome’s own PDF engine instead of screenshots

Tools that build the PDF from a screenshot turn the page into a picture, so the text can’t be searched or copied and images lose sharpness. This extension drives the same engine Chrome uses for printing, which keeps real text and the original image files. The cost is a “started debugging” bar that Chrome shows for a few seconds during each save. The popup says so up front instead of letting it surprise people.

Automatic removal with a manual override

Rules for spotting navigation and popups handle most sites, but no rule set is right everywhere. The click-to-remove picker lets people fix the rest in a few clicks instead of giving up on the page.

Paper pages by default

The first version produced one tall page that matched the screen. Testing showed those PDFs were hard to read and print, so the default changed to normal paper pages. Saved settings from the old default are migrated automatically.

As few permissions as possible

The extension doesn’t ask for access to any websites up front. It only works on a tab when you save it, and it never sends page content anywhere. That matters for a tool that reads every page you save.

Errors that say what to do next

Pages Chrome doesn’t let extensions touch, such as its settings pages, get a plain explanation. A tab that is already a PDF gets a “Download original” button instead of an error.

What changed after testing

I tested each version on the real pages I needed it for and turned what broke into the next round of work. Every fix came with automated tests so the problem can’t quietly come back.

Round	What I saw	Cause	What changed
1	A course lesson didn’t scroll, was cut off, and the course outline stayed in the PDF	The lesson runs inside frames nested two levels deep in a full-screen player	Prepare every frame on the page, grow each one to its full height, and print the player on its own
2	A blank first page, an image cut across two pages, and pages far too long to read	The page behind the player printed first, and the default was one long page	Paper pages by default, images and slides kept whole at page breaks, lesson starts on page 1
3	A PDF that was one big blank page	Hiding an empty outline column collapsed the lesson to zero width	Fixed the layout step and added a check that no page is empty
4	Error entries on Chrome’s extensions page	A normal refusal was logged as an error, and the popup lost its connection when Chrome paused the extension after 30 idle seconds	Expected refusals are no longer logged, and the popup reconnects on its own
5	Every lesson got the same file name, and the save dialog always opened in the default folder	The page title names the module rather than the lesson, and Chrome sends extension-named downloads to the default folder	Files are named after the lesson heading, and saves go through Chrome’s normal download path so it remembers the folder
6	A lesson’s image carousel came out as its first image only	The other slides sat beside the first one, cut off by the carousel’s box and marked hidden, and the last two images never loaded	Carousels are laid out so every slide prints in order, loop copies and arrows are left out, and every slide’s image loads

How quality is checked

The original requirements became automated tests: 50 in total, 31 of which load the extension in a real Chrome and inspect the PDFs it produces. They cover these requirements:

• Text is real text. Fonts are embedded and the page text can be extracted from the PDF.
• Images keep their quality. Each embedded image has its source’s full resolution, and JPEG files are embedded byte for byte.
• Removal works. Navigation, cookie banner and chat text is missing from the PDF while the article text is there.
• The whole page loads. The last batch of an infinite feed makes it into the PDF.
• Page breaks are clean, with no blank pages and no slide split across two pages.
• Carousels print every slide once, in order, with each slide’s image and without arrows or dots.

Limitations

• Apps that draw on a canvas, such as Google Docs, Sheets and Figma, come out as images rather than text.
• Feeds that unload items as you scroll past them, such as X, Slack and Gmail, only include what’s on screen.
• Carousels that only add a slide to the page when you move to it print just the slide that’s showing.
• Chrome shows its “started debugging” bar for a few seconds during each save.

The full list is in How it works.

Ideas for what’s next

• Publish it on the Chrome Web Store so it installs in one click.
• Save a whole course, or a set of open tabs, in one go.
• Remember what you removed on a site and apply it next time.

My role

I defined the problem and the requirements, chose between the options at each decision point, tested every version on real pages, and decided what to fix next. The code was written with Claude Code, an AI coding assistant, working from those requirements and my test feedback. It was built over two days in October 2026.

Try it

Requires Chrome 125 or newer.

1. Download this repository (Code > Download ZIP, then unzip it) or clone it.
2. Open chrome://extensions.
3. Turn on Developer mode (top right).
4. Click Load unpacked and choose the downloaded folder (pagevera-main if you used Download ZIP).
5. Optional: pin the extension from the puzzle-piece menu so the button is always visible.

To save local file:// pages, open the extension’s Details and turn on Allow access to file URLs.

Then click the toolbar button and choose Save as PDF.

• Press Alt+Shift+P (⌥⇧P on a Mac) to save without opening the popup. Change it at chrome://extensions/shortcuts.
• Right-click a page and choose Save page with PageVera. It uses the format you last picked in the popup.
• Pick what to remove first… puts a toolbar on the page. Hover to highlight an element, click to remove it, ↑/↓ to select the parent or child, Backspace or ⌘Z to undo, Enter to save, Esc to cancel. Show automatic removals outlines what the Remove toggles would take out; click an outline to keep that element. After saving, the removed elements stay hidden until you click Restore page or reload, so you can adjust and save again.

While a PDF is being made, Chrome shows a “PageVera started debugging this browser” bar for a few seconds. It disappears when the save finishes. Clicking its Cancel button stops the save. To hide the bar permanently, start Chrome with --silent-debugger-extension-api.

Settings

All settings are in the popup and are remembered.

Option	Default	What it does
Top navigation and header	on	Removes site headers, nav bars and announcement bars at the top, including sticky and fixed ones
Side navigation and sidebars	on	Removes left and right columns such as doc sidebars and “On this page” lists, then lets the content use the freed space
Popups, cookie banners, chat widgets	on	Removes consent banners, modals, chat launchers and bottom bars, and undoes scroll locks
Footer	off	Removes the site footer
Scroll to the end	on, 60 s	Scrolls through the page so lazy images and infinite feeds load, then returns to the top
Save as	PDF	Markdown saves the page’s text as a .md file, with headings, lists, tables, code blocks and links kept and the same removals applied. Images become links to the original files, so they show only while you’re online and the site still serves them. Content inside embedded players comes after the main page’s text. PNG image saves one picture of the whole page at its on-screen width, with the same removals and scrolling. Text in a PNG can’t be searched or copied, and the layout options don’t apply. Pages taller than 16,384 px at 2× are saved at a lower pixel ratio.
Layout	Pages, Auto size, small margins	Normal pages with the desktop layout scaled to fit the paper width. Auto size picks Letter in the US and similar regions and A4 elsewhere. Images, slides and lines of text are kept whole at page breaks, including inside embedded course players. One long page gives one tall page at your screen width instead; pages taller than 200 inches (the PDF viewer limit) are split into equal parts.
Load the highest-resolution images	on, 2×	Switches responsive images (srcset, <picture>, image-set()) to their largest version
Keep the on-screen look	on	Ignores the site’s print stylesheet
Stop animations	on	Finishes fade-in and slide-in animations so nothing prints half-transparent
Bookmarks and accessibility tags	on	Adds PDF bookmarks from headings and a tagged structure. Turned off automatically for very large pages.
Ask where to save	off	Shows the Save dialog even when Chrome’s own “Ask where to save each file” setting is off. Chrome opens this dialog in the default download folder, not the last one used, so turn on Chrome’s setting instead if you want that.
Downloads subfolder	none	Saves into a folder inside the default download folder

More

• How it works: the save pipeline, Chrome’s PDF engine, and how to run the tests.
• License: MIT.

This site is open source. Improve this page.