Free image downloader and screenshot extension for Chrome and Edge. Bulk download images, save image as PNG, JPG or WebP, and take full page screenshots. No accounts, no tracking.
Install · Features · Docs · Changelog · Issues
Applies to: ImgFetch 1.1.0 Last verified: 2026-10-22
Version 1.1.1, released on GitHub Releases on 2026-10-25 (tag v1.1.1). Version 1.1.0 was released on 2026-10-22 and version 1.0.0 on 2026-10-12.
| Browser | Install from | Status |
|---|---|---|
![]() |
Edge Add-ons | Published. Version 1.1.1 is live. |
![]() |
Latest release on GitHub | Chrome Web Store listing coming soon. For now, download the ZIP from the latest GitHub release and install it unpacked (steps below). |
![]() |
Firefox Add-ons | Coming soon (needs a separate port) |
| GitHub Releases | Latest: 1.1.1, ZIP download for Chrome |
The Chrome Web Store listing is coming soon. Until then, install ImgFetch in Chrome from the release file:
- Download the ZIP from the latest GitHub release and unzip it. The unzipped folder must contain
manifest.jsonat the top level. - Open
chrome://extensionsin Chrome. - Turn on Developer mode (top right corner).
- Click Load unpacked and select the unzipped folder. You can also drag the unzipped folder onto the page.
- Pin ImgFetch from the extensions menu if you want it on the toolbar.
Chrome may show a developer mode notice for extensions installed this way, and it does not update them automatically. To update, download the newest release and repeat these steps.
You can install the official extension directly from the Microsoft Edge Add-ons store. Note: The current live listing serves version 1.1.1.
You can also clone the repository with git clone https://github.com/stillware/ImgFetch and load the cloned folder with Load unpacked the same way.
- Install
- Features
- See it in action
- Quick start
- Keyboard shortcuts
- Privacy and permissions
- Supported browsers and languages
- Architecture
- Documentation
- Contributing
- Security
- Translations
- Release history
- FAQ
- License
- About
ImgFetch is a free, open-source image downloader and screenshot extension for Google Chrome and Microsoft Edge. It finds the real sources behind a page image, checks that each source loads, and saves the best valid one. It never upscales an image and never claims quality it has not verified.
| Area | What it does |
|---|---|
| Image downloads | Right-click Smart Download saves the best validated source. It never upscales and never invents a better source. |
| Whole page scan | Scan This Page finds all images across a whole page for preview, filtering and bulk image download. |
| Bulk and ZIP | Download a selection one by one, or pack it into a single ZIP file. |
| Formats and conversion | Save image as PNG, JPG, WebP, AVIF, GIF, SVG or BMP (for example WebP to PNG or JPG) where the browser supports the format. Quality settings apply per format, and PNG at full quality is lossless. |
| Screenshots | Take a full page screenshot (scrolling capture), or capture the Visible area, a selected Area or a single Element. Long pages are split into balanced parts and reviewed in the Parts overview. |
| Editor | Annotate screenshots with Pen, Arrow, Rectangle, Circle, Text and Pixelate, adjust the frame with the non-destructive Crop tool, then use Undo, Redo, Copy and Save. |
| Languages | The interface is available in 21 locales covering 19 languages, with a right-to-left layout for Arabic. |
| Privacy | Everything runs on your device. There are no accounts, no analytics, no telemetry and no tracking. |
- Open a page with images and click the ImgFetch toolbar button to scan the page.
- Preview the results, filter if needed, and download one image or a selection.
- Or right-click any image and choose Smart Download to save its best source at once.
- Use the popup screenshot buttons to capture the Visible area, an Area, an Element or the Full page.
- Annotate or Crop the screenshot in the editor, then choose Save or Copy.
Defaults come from the extension manifest.
You can change them at chrome://extensions/shortcuts in Chrome or edge://extensions/shortcuts in Edge.
On macOS, use Command instead of Ctrl for the extension shortcuts.
| Command | Default key | Action |
|---|---|---|
| Scan page | Ctrl+Shift+S | Scan the page for images |
| Scan page (alternative) | No default key | Scan the page for images; bind your own key in the browser shortcut settings |
| Capture area | Ctrl+Shift+X | Capture a selected Area |
| Capture element | Ctrl+Shift+E | Capture a single Element |
| Select element | Alt+Shift+E | Pick an Element and show its image details |
Editor shortcuts, active in the screenshot editor:
| Shortcut | Action |
|---|---|
| C | Select the Crop tool |
| Ctrl+Z (Cmd+Z on macOS) | Undo |
| Ctrl+Y | Redo |
| Ctrl+Shift+Z (Cmd+Shift+Z on macOS) | Redo |
| Enter | Apply a pending crop, or confirm a Text label |
| Esc | Cancel a pending crop or a Text label |
| Arrow keys | Nudge a pending crop by 1 pixel (Shift+arrow keys: 10 pixels) |
| Delete or Backspace | Delete the selected Text label |
Undo does exactly what the Undo button does, including the crop step in the same history, and it does not fire while you type in the text annotation box, where the browser native text undo keeps working.
The full policy is in PRIVACY.md, and docs/permissions.md explains each permission.
- No accounts: ImgFetch has no sign up, no sign in and no user profiles.
- No tracking: there is no analytics, no telemetry, no advertising and no data collection of any kind.
- No remote code: the extension runs only the code shipped in the package and loads no scripts from the network.
- Permissions are used only for the feature they support, such as downloads for saving files, scripting for scanning pages and clipboardWrite for the Copy actions; docs/permissions.md lists each one.
- Your data stays on your device: settings, download history and captures are stored locally in your browser and are never sent anywhere by ImgFetch.
| Browser | Support |
|---|---|
| Google Chrome | Supported; requires Chrome 116 or later |
| Microsoft Edge | Supported; Chromium based Edge |
| Other Chromium browsers | Not officially tested; the extension may load and work |
| Firefox | No Firefox version exists yet; a Firefox version is coming soon and needs a separate port |
The interface is available in 21 locales covering 19 languages: English, Spanish, Latin American Spanish, Brazilian Portuguese, French, German, Italian, Dutch, Russian, Ukrainian, Polish, Czech, Turkish, Arabic, Simplified Chinese, Traditional Chinese, Japanese, Korean, Indonesian, Vietnamese and Thai. Change it under Options, General, Language; the default is Auto, which follows the browser language. Arabic uses a right-to-left layout. Translations are AI assisted and have not been reviewed by native speakers; corrections are welcome through an issue or docs/translating.md.
ImgFetch runs entirely on your device. The popup, the options page and the page scripts send messages to one service worker. The worker validates each message, keeps its decision logic in pure core modules, hands heavy jobs to the offscreen document, and uses the browser services to save files and data. The screenshot pages read their captures from IndexedDB.
%%{init: {"flowchart": {"curve": "basis", "nodeSpacing": 24, "rankSpacing": 56, "padding": 14}}}%%
flowchart LR
user(["<b>User</b>"])
subgraph g_ui["Interface"]
popup["<b>Popup</b><br/>popup.js"]
options["<b>Options</b><br/>options.js"]
end
subgraph g_page["Inside the web page"]
scanner["<b>Page scanner</b><br/>scanner.js"]
overlay["<b>Area and Element picker</b><br/>overlay.js"]
end
subgraph g_shots["Screenshot pages"]
capture["<b>Capture page</b><br/>capture.js"]
parts["<b>Parts overview</b><br/>parts.js"]
annotate["<b>Editor</b><br/>annotate.js"]
captureio["<b>Capture access</b><br/>capture-io.js"]
copyshot["<b>Copy screenshot</b><br/>copy-shot.js"]
end
sw{{"<b>Service worker</b><br/>service-worker.js<br/>validates every message,<br/>routes all work"}}
subgraph g_core["Core logic (pure, no chrome.* calls)"]
rules["<b>Rules and safety</b><br/>validation.js, urls.js,<br/>filenames.js"]
state["<b>Queue, settings, history</b><br/>queue.js, settings.js,<br/>history.js"]
plan["<b>Capture planning</b><br/>fullpage-capture.js,<br/>capture-storage.js"]
end
subgraph g_off["Offscreen document"]
offscreen["<b>Heavy work</b><br/>format conversion, SVG,<br/>ZIP (zip.js), stitching,<br/>clipboard<br/>offscreen.js"]
end
subgraph g_browser["Browser services"]
downloads["<b>Downloads API</b><br/>saves files"]
storage[("<b>Browser storage</b><br/>settings, history")]
idb[("<b>IndexedDB</b><br/>captures, idb.js")]
end
user -->|"scans"| popup
user -->|"settings"| options
user -->|"browses"| scanner
user -->|"captures"| capture
popup -->|"messages"| sw
options -->|"messages"| sw
scanner <-->|"sources"| sw
overlay -->|"area"| sw
scanner --- overlay
parts -->|"save"| sw
sw -->|"validates"| rules
sw -->|"tracks"| state
sw -->|"plans"| plan
sw -->|"convert, ZIP"| offscreen
sw -->|"downloads"| downloads
sw -->|"persists"| storage
sw -->|"captures"| idb
capture -->|"reads"| captureio
parts -->|"reads"| captureio
annotate -->|"reads"| captureio
captureio -.->|"loads"| idb
parts -->|"editor"| annotate
annotate -->|"Copy"| copyshot
classDef actor fill:#eef0ff,stroke:#5b50e6,stroke-width:2px,color:#171a2e
classDef ui fill:#e0e7ff,stroke:#5457f3,stroke-width:1.5px,color:#171a2e
classDef page fill:#dbeafe,stroke:#3d6cf9,stroke-width:1.5px,color:#171a2e
classDef brain fill:#ddd6fe,stroke:#6d28d9,stroke-width:3px,color:#171a2e
classDef pure fill:#f3e8ff,stroke:#a855f7,stroke-width:1.5px,color:#171a2e
classDef work fill:#cffafe,stroke:#0891b2,stroke-width:1.5px,color:#171a2e
classDef shot fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#171a2e
classDef store fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#171a2e
class user actor
class popup,options ui
class scanner,overlay page
class sw brain
class rules,state,plan pure
class offscreen work
class capture,parts,annotate,captureio,copyshot shot
class downloads,storage,idb store
style g_ui fill:none,stroke:#5457f3,stroke-width:1.5px,stroke-dasharray:6 4
style g_page fill:none,stroke:#3d6cf9,stroke-width:1.5px,stroke-dasharray:6 4
style g_core fill:none,stroke:#a855f7,stroke-width:1.5px,stroke-dasharray:6 4
style g_off fill:none,stroke:#0891b2,stroke-width:1.5px,stroke-dasharray:6 4
style g_shots fill:none,stroke:#16a34a,stroke-width:1.5px,stroke-dasharray:6 4
style g_browser fill:none,stroke:#d97706,stroke-width:1.5px,stroke-dasharray:6 4
linkStyle default stroke:#6c63ff,stroke-width:1.5px
Solid arrows are messages and calls. Dotted arrows are direct reads.
| Step | What happens |
|---|---|
| 1 | You select an image in the popup (or use the right-click menu) and start a download. |
| 2 | The popup sends a message. The service worker checks the message and its sender with validation.js before it acts. |
| 3 | The worker asks the page scanner for the ranked, validated sources of that image and gets the best one first. |
| 4 | If you chose another format, the offscreen document converts the image, or builds the ZIP with zip.js. |
| 5 | The worker starts the download through the browser Downloads API with a safe file name from filenames.js. |
| 6 | When the download completes, the worker writes a history record and keeps the queue state in browser storage. |
Screenshots follow the same pattern. The picker returns the area or element you chose, the worker captures the page and stores the result in IndexedDB, and the screenshot pages (Capture, Parts overview, Editor) read it from there through capture-io.js.
| Color | Layer | Folders |
|---|---|---|
| Indigo | What you see and click | popup/, options/ |
| Blue | Runs inside the web page | content/ |
| Violet (hexagon) | Decides and routes | background/service-worker.js |
| Purple | Pure logic, tested without a browser | core/ |
| Cyan | Heavy work off the worker | offscreen/ |
| Green | Screenshot result pages | pages/, shared/ |
| Amber | Browser services and storage | Downloads API, browser storage, IndexedDB |
- Pure core, thin edges. Decision logic lives in
core/with nochrome.*calls, so Node unit tests run it directly. - Validate, then route. Every runtime message is checked before the service worker acts on it.
- Fail closed. Unknown message types, unknown senders and malformed payloads are denied, never guessed at.
- No remote code and no build step. The extension ships plain JavaScript, HTML and CSS.
For the full component list, message protocol and sequence diagrams, see docs/architecture.md.
Root documents:
- CHANGELOG.md: what changed in each version.
- CONTRIBUTING.md: how to set up, test and submit changes.
- SECURITY.md: how to report a vulnerability.
- CODE_OF_CONDUCT.md: the standard expected in project spaces.
- PRIVACY.md: the full privacy policy.
- TERMS.md: terms of use.
Guides and references in docs/:
- docs/README.md: the documentation index with reading paths.
- docs/user-guide.md: task based guide for every feature.
- docs/architecture.md: components, messages, storage and design.
- docs/project-structure.md: annotated file tree and naming rules.
- docs/permissions.md: every permission, why it exists and what happens without it.
- docs/security-model.md: threat model and mitigations.
- docs/limitations.md: known limits, reasons and workarounds.
- docs/performance.md: measured performance and design notes.
- docs/build.md: how to build, check and package the extension.
- docs/translating.md: how translations are organized and how to add one.
- docs/glossary.md: short definitions of terms used in the code and docs.
- docs/tests/README.md: test suites, how to run them and what they cover.
- docs/tests/manual-test-checklist.md: checks that need a human and a real browser.
- docs/decisions/README.md: index of architecture decision records.
- docs/history/README.md: index of historical build records.
- store-assets/store-listing.md: store listing text and the pre-submission checklist.
The code is split by role: background/ holds the service worker, content/ holds the page scanner and overlay, core/ holds pure logic modules, offscreen/ holds the hidden document used for conversion and stitching, and popup/, options/ and pages/ hold the user interface.
Shared interface code lives in shared/, tests in tests/ and build tooling in scripts/.
The full annotated tree is in docs/project-structure.md.
Contributions are welcome; start with CONTRIBUTING.md and open an issue before large changes so the approach can be agreed first.
Run the test suites and scripts/validate.py before opening a pull request.
Project spaces follow the Code of Conduct.
To report a security problem, follow SECURITY.md and use GitHub private vulnerability reporting rather than a public issue.
Translations live in _locales/, one folder per locale. See docs/translating.md for how to add or fix a language, and run node scripts/check-i18n.mjs before submitting.
- 1.1.1, 2026-10-25: the current release. Link fixes and a short README inside the ZIP. See CHANGELOG.md.
- 1.1.0, 2026-10-22: the previous release. See CHANGELOG.md for what changed and GitHub Releases for the downloads.
- 1.0.0, 2026-10-12: the first public release.
1. Is ImgFetch free?
Yes. ImgFetch is free software under the GNU General Public License v3 or later, with no paid tier and no accounts.
2. Does it upload my images anywhere?
No. Downloads, conversions and captures happen on your device. The only network requests are the ones needed to fetch images you chose to save.
3. Why is the Chrome Web Store listing coming soon?
The Chrome Web Store listing is not live yet; until it is, install 1.1.0 from GitHub Releases as described under Install.
4. Why is AVIF sometimes unavailable?
AVIF encoding depends on the browser, and ImgFetch offers it only when a real encode test succeeds. On older browsers it shows a clear message instead of silently saving a different format.
5. Why does a copied screenshot not appear in Windows clipboard history?
On Windows, clipboard history (Win+V) does not show images above roughly 4 MB. The copy still worked: paste it with Ctrl+V into Paint, a document or a chat. ImgFetch shows a short note in the editor for large images on Windows, and a longer copy toast for very large copies, so this limit does not look like a failed copy.
6. What happens when a full page is too long for one screenshot?
The capture is split into balanced parts, each up to 32767 pixels per side and about 100 megapixels, stored on your device. When the capture finishes, one Parts overview page opens where you can review each part and Edit, Save, Copy or Discard it, or Save all parts at once. Parts already finished are kept if you stop the capture or switch tabs.
7. How does Crop work in the editor?
Crop is non-destructive: the original image and your annotations are never modified, and the crop rectangle is stored alongside them in the same Undo and Redo history. Drag the rectangle, resize it with its 8 handles, nudge it with the arrow keys, then apply it with Enter. Save and Copy export exactly the cropped region at full quality.
ImgFetch is free software: you can redistribute it and modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. See LICENSE for the full text.
The GPL covers the source code. It does not grant the right to use the ImgFetch name or logo as branding for a modified version. See TRADEMARK.md.
Official sources: this repository, its GitHub Releases and the Microsoft Edge Add-ons listing. A copy that uses the ImgFetch name from anywhere else is not official.
ImgFetch is free and open source software with no accounts, no tracking and all processing on your device. Copyright (C) 2026 ImgFetch Developer.











