Comparison

Image extractor extension permissions, compared

Every image downloader asks for something. Some ask to read the tab you are looking at; others ask to read every page you visit, forever. This page sets out what each one actually requests - read from published source, with a link to the evidence for every row.

Verified 10 August 2026

The short version

  • Most image extraction extensions request <all_urls> - permanent read and write access to every website you visit.
  • Images Extractor requests activeTab instead: the one tab you are looking at, only at the moment you click the icon.
  • Broad access is not automatically sinister. It is usually the price of bulk features - and this page says plainly where that trade-off favours the others.
  • You can check any of this yourself in about a minute. We show you how.

The comparison

Extension MV Site access Other notable permissions Evidence
Images Extractor
(this one)		 3 activeTab
Current tab, on click only 		scripting, downloads, clipboardWrite manifest.json
Image Downloader — all site data
Per developer's own justification 		Not published Web Store listing
Mass Image Downloader 3 <all_urls>
Host access + content script on every page 		tabs, storage, clipboardRead, downloads, scripting manifest.json
Pic-Grabber 3 <all_urls>
Host access + content script on every page 		webRequest, tabs, sidePanel, storage, downloads manifest.json
Media Downloader 3 <all_urls>
Host access 		sidePanel, contextMenus, storage, tabs, downloads manifest.json
Image Downloader Continued 2 <all_urls>
Listed directly in permissions 		webNavigation, downloads, activeTab manifest.json
Download All Images — not published Not published Web Store listing

Rows marked not published are exactly that - the extension's source is not public and its Web Store listing does not enumerate permissions. We have left those cells empty rather than guess. See how this was verified.

What activeTab actually means

The distinction that matters here is not how many permissions appear in a list, but whether an extension can read pages you never asked it to touch.

activeTab grants access to a single tab, and only when you click the extension's icon. It expires when you navigate away. The extension cannot see the tab before you click, cannot see your other tabs, and cannot run anything in the background.

<all_urls> in host_permissions is standing access to every site, granted once at install and lasting until you remove the extension. Paired with a content script matching <all_urls>, code runs on every page you load whether or not you ever open the extension again. Chrome reflects this at install time: "Read and change all your data on all websites."

Two further permissions in the table are worth naming. clipboardRead allows an extension to read what is on your clipboard, which is a different and larger ask than clipboardWrite - the write-only permission Images Extractor uses. And webRequest allows observing your network traffic.

In fairness to the others

A broad permission is not evidence of bad intent, and nothing on this page should be read that way. Every extension listed above discloses no data collection, and several are open source - which is precisely why their manifests could be quoted here at all. That openness deserves credit, not suspicion.

Broad access is usually the honest cost of broader features. Scanning several tabs at once, crawling paginated galleries, filtering by dimension, or zipping a hundred files genuinely cannot be done from a single-tab, click-scoped permission. Those extensions ask for more because they do more.

The point of this table is not that one design is virtuous and another is not. It is that the trade-off should be visible before you install, and right now it usually is not.

What Images Extractor gives up for this

A comparison that only flatters the author is not worth reading. The narrow permission model has real costs, and here they are:

  • No bulk download. There is no "download all as ZIP". You save images one at a time. If you routinely need fifty files at once, one of the extensions above will serve you better.
  • One tab at a time. No multi-tab scanning, no gallery crawling, no following pagination.
  • No filtering or sorting. You get every image the page exposes, in document order - no minimum-size filter, no type-based exclusion.
  • Some images cannot be copied to the clipboard. Without host permissions, reading an image's bytes depends on the host allowing it. Where it does not, the copy-image action reports that plainly and the URL, download, and open actions still work. This is a direct consequence of not requesting broad access.

If those limits do not affect how you work, the narrower permission footprint is free. If they do, install something else - that is a legitimate choice, and knowing why is the entire point.

Check any extension yourself in a minute

You do not have to trust this table. Both methods below work on any extension you already have installed.

From Chrome's own UI

  1. Open chrome://extensions/.
  2. Find the extension and click Details.
  3. Read the Permissions and Site access sections. "On all sites" means standing access to everything you browse.

From the extension's own manifest

Every installed extension keeps its manifest.json on disk in plain text. On macOS, look under ~/Library/Application Support/Google/Chrome/Default/Extensions/, then the extension's ID and version folder. Open manifest.json and read the permissions, host_permissions, and content_scripts keys. That file is the authority - it is what Chrome itself enforces, and it cannot be spun by a store description.

How this was verified

Every claim above comes from one of two sources, and the table says which for each row:

  1. A published manifest. For the open-source extensions, the permissions, host_permissions, and content_scripts keys were read directly from manifest.json in the project's own repository. This is authoritative - it is the same file Chrome enforces.
  2. The developer's own Web Store disclosure. Where source is not public, only what the developer states on their listing is reported. Image Downloader's listing justifies its access this way:
    "To extract photos from a webpage, Image Downloader must access all content when the popup is activated. No data is sent off your device."

Known limitations, stated plainly: this is not an exhaustive survey of every image extension in the Web Store. It covers those whose permissions could be established from primary sources on the verification date. Extensions update, so a manifest read today may differ tomorrow - the date at the top of this page is when each row was last checked, and the evidence links let you re-check any of them yourself.

Found a row that is wrong or out of date? Open an issue and it will be corrected. Accuracy here matters more than the comparison being favourable.

Fewer permissions, same job

Extract, copy, and download every image on a page - without granting access to every page you visit.

⬇ Download Images Extractor