Image Search v3

AnkiWeb addon 178037783

Fork of Anki Image Search v2 Addon (AI-detected; may be wrong)

2 other versions of this addon exist. Compare versions ↓

Searches multiple image providers using note fields or selected text and inserts chosen images into a configured field, with browsing, per-note-type settings, and smart replacement.
AI-generated summary; may contain mistakes.

card-creationfork-or-reupload

Open on AnkiWeb GitHub Forum topic Find alternatives

AnkiWeb

Votes
3 (👍 2 · 👎 1)
Updated
2026-10-16
Anki versions
26.05~
License
GPL-2.0 (from the GitHub repository)
Virus scan
No ClamAV match A match does not prove malware. No match does not prove that the code is safe.
Languages
Python, JSON
Description language
en

Maintenance

active

  • Last update or commit: 2026-10-16 (16 days before the snapshot).
  • Test files: 12.
  • GitHub Actions workflows: 1.
  • Commits in the last year: 45.
  • Issues: 1 open, 2 closed.
  • Pull requests: 0 open, 0 merged.
  • Median first response: 473 hours (3 answered issues and pull requests).

Will it work on my Anki?

Version branches
Min AnkiMax AnkiUpdated
2.1.126.05+2026-10-16
History across monthly snapshots

Loading…

Compare versions

The original addon and its forks. The best value in each column is in bold.

VersionWorks on Anki 26.09.3MaintenanceLast activity VotesApprovalGitHub starsNotes
Image Search original unknown abandoned 23 Jan 2018 👍 12 · 👎 17 26% 10
↳ Anki Image Search v2 Addon likely abandoned 21 Aug 2022 👍 11 · 👎 5 44% 1
↳ Image Search v3 this addon (AI-detected; may be wrong) likely active 16 Oct 2026 👍 2 · 👎 1 21% 2

Forks can add features or work in a different way. Read the description before you change to another version. Fork detection is incomplete, so other versions can exist.

Similar addons

Anki Image Picker

Lets you search multiple providers and interactively choose thumbnails before inserting images into selected Anki notes, with batch processing and session controls.

card-creationnote-managementshortcuts

5 votes 57% approval 👍 5 · 👎 0 ⭐ 0 Anki 26.12~ Updated 2026-04-03
Cambridge/Wiktionary fetcher

Fills Anki notes with dictionary data—definitions, examples, synonyms, part of speech, IPA, audio, and images—from Cambridge and Wiktionary sources, using configurable field mapping and typo suggestions.

card-creationdictionary

2 votes 34% approval 👍 2 · 👎 0 ⭐ 1 Anki 2.1.1~ Updated 2026-03-23
PDF to notes has 1 fork →

Converts PDF pages into Anki normal or cloze notes, using each page’s first line as prompt and page images or HTML, with multiple-file, subdeck, scaling, and shortcut options.

card-creationimportcloze

9 votes 27% approval 👍 5 · 👎 4 ⭐ 6 Anki 2.1.66~ Updated 2022-07-24

Description

Image Search v3

Install from ankiweb

Image Search v3 is a powerful Anki add-on that lets you quickly find and add images to your cards directly from the editor. It searches for images based on the content of your fields or selected text and places the chosen image into a designated field.

Features

  • Image provider: Yandex (default, no key), Bing (no key), DuckDuckGo (hidden API, no key), Brave Search API (requires subscription token), or Google Custom Search (images). When using Google, enter your API key and CSE ID (cx) under Tools → Image Search v3 Settings → Network. [Requires both key and cx] [searchType=image]. When using Brave, enter your subscription token under Network. Configure a global ordered fallback list to use any provider after any other provider returns no results or errors.
  • Per-Note-Type Configuration: Configure different query and image fields for each of your note types.
  • Smart replace: only replaces prior images inserted by this add‑on (class "imgsearch"), preserving user text and other content; appends when no prior add‑on image exists.
  • Multiple images per card: search a word, paste its image, then search another word — the new query appends its image instead of overwriting the previous one, so one card can hold an image per searched term. Browsing (⬅ ➡) or re‑searching the same query still swaps the last image in place, keeping editor focus and undo history.
  • Graphical Settings Panel: An easy-to-use settings panel to manage your configuration. No more manual file editing!
  • Smart Defaults: Automatically uses the first field of a note type for searching and the last field for placing the image if not configured otherwise.
  • Search on Selection: Simply highlight any text in the editor and use the search button or right-click context menu to search for an image.
  • Toolbar Integration: Adds 🖼, ⬅, and ➡ buttons directly to the Anki editor toolbar for a fast workflow.
  • Right-Click Context Menu: Right-click on highlighted text to instantly start an image search.
  • Support Tab with Supporter Opt-out: A dedicated Support tab with Ko-fi, UPI, BTC, and ETH addresses, and an "I have supported this addon" checkbox that hides the automatic post-update welcome.
  • Update Welcome: When the add-on is updated, the Support tab opens automatically the next time you open the settings dialog (no startup cost — the check runs only on dialog open). The opt-out checkbox in the Support tab turns this off.
  • Built-in Log Viewer: The Logs tab shows a rotating log file (512 KiB × 3). Lazy-loaded — nothing is read from disk until you click Refresh. A "Log debug" checkbox toggles maximum verbosity on/off without picking a level from the dropdown; the "All" level captures even sub-DEBUG noise from third-party libraries.
  • Log Health Scan: A "Check log file" button on the Logs tab scans the log for known error patterns (tracebacks, timeouts, permission errors, provider give-ups, etc.) and shows a per-category summary so you can spot issues without re-reading the whole file.
  • Non-Modal Settings: The settings dialog is non-modal — you can keep editing cards in Anki while it's open. A second click on the menu re-focuses the existing window instead of stacking a new one.

Usage

1. Configuration

  • Open Tools → Image Search v3 Settings, select a note type on the left, then configure Query Fields, the Image Field, and Image Placement on the right, and click Save.
  • Under Network, pick the provider; for Google, paste your API key and CSE ID (cx), and adjust timeout, retries, and backoff to suit your network.
  • Defaults ship with an empty key/cx and provider set to Yandex, and sane network values for reliable startup on new installs.
Screenshot_20251104_012633 Screenshot_20251104_020651

2. Searching for Images

There are three ways to search for an image in the card editor:

  1. Using the Toolbar Button: Click the 🖼 button on the editor toolbar.
    • If you have text highlighted anywhere in the editor, that text will be used for the search.
    • If no text is highlighted, the content of your configured Query Field(s) will be used.
  2. Using the Right-Click Menu: Highlight the text you want to search for, right-click it, and select "Search image for: '...'" from the context menu.
  3. Browsing Results: Use the ⬅ and ➡ buttons to browse through other image results for the last query that was performed from the query field(s).
Screenshot_20251031_152224 Screenshot_20251031_152301 Screenshot_20251031_152339

3. Support tab & post-update welcome

  • Open Tools → Image Search v3 Settings → Support to find Ko-fi, UPI, BTC, and ETH addresses.
  • If you have supported the add-on, tick "I have supported this addon" at the top of the Support tab. The add-on records your preference in meta.json (not in your committed config) and stops opening the Support tab on every update.
  • The post-update welcome only runs when you open the settings dialog (not at Anki startup), so it has zero effect on startup time.
  • To disable the welcome entirely (for example in a managed environment), set auto_show_support_on_update to false in your add-on config (the Logs tab is the easiest place to edit JSON via Tools → Image Search v3 Settings → Logs; or use Tools → Add-ons → Image Search v3 → Config).

Provider notes

  • Yandex: no‑auth, undocumented JSON endpoint used by the front‑end; works well but may change, be geo‑restricted, or rate‑limited without prior notice.
  • Bing: no‑auth, undocumented /images/async endpoint that returns ~30–50 results per query. No JavaScript rendering required; we extract the murl field from the response HTML.
  • Brave: first-party Image Search API. Requires a paid plan and the X-Subscription-Token header; we use the properties.url (original) or thumbnail.src (Brave-proxied) fields from the response. Up to 200 results per request.
  • DuckDuckGo: hidden i.js endpoint (no API key); works best‑effort and may change, rate‑limit, or block without notice. A Linux Chrome User-Agent is required because DDG returns 403 for some Windows UAs. Uses the configured global fallback chain if no results.
  • Google: official Custom Search JSON API with searchType=image; requires both API key and CSE (Google Search Engine) ID (cx) and enforces quotas and billing on your account.
  • Routing: the selected provider is tried first, then the configured global fallback providers in order, preserving the editing flow when any provider returns no results or errors.

If you don't know how to get the API please read this: google custom-search

Brave API key: https://brave.com/search/api/

Developer documentation lives under doc/ (architecture, modules, JSON keys, errors, build, security, testing).

Troubleshooting

Context Menu Item Not Appearing

If the "Search image for..." option does not appear when you right-click on selected text, it might be due to a conflict with another add-on that also modifies the context menu. A common conflict is with add-ons that provide image editing or other right-click functionalities in the editor.

You can diagnose this by temporarily disabling other editor-related add-ons (like "Image Editor") via Tools -> Add-ons, restarting Anki, and checking if the menu item appears.

Support

If you find this add-on useful, please consider supporting its development:

ko-fi

Changelog

The full release history is in CHANGELOG.md (newest first, grouped by version, with links to the corresponding GitHub releases). Contributor and maintainer documentation (architecture, modules, JSON keys, errors, build, security, testing) lives under doc/.

License

This add-on is a modification of the work of original authors. Credit goes to the creators of Anki Image Search v2 and Image Search.

Toolbar buttons use emoji labels to avoid bundling image assets.

This project is licensed under the GPLv2.

README

Image Search v3

Install from ankiweb

Image Search v3 is a powerful Anki add-on that lets you quickly find and add images to your cards directly from the editor. It searches for images based on the content of your fields or selected text and places the chosen image into a designated field.

Features

  • Image provider: Yandex (default, no key), Bing (no key), DuckDuckGo (hidden API, no key), Brave Search API (requires subscription token), or Google Custom Search (images). When using Google, enter your API key and CSE ID (cx) under Tools → Image Search v3 Settings → Network. [Requires both key and cx] [searchType=image]. When using Brave, enter your subscription token under Network. Configure a global ordered fallback list to use any provider after any other provider returns no results or errors.
  • Per-Note-Type Configuration: Configure different query and image fields for each of your note types.
  • Smart replace: only replaces prior images inserted by this add‑on (class "imgsearch"), preserving user text and other content; appends when no prior add‑on image exists.
  • Multiple images per card: search a word, paste its image, then search another word — the new query appends its image instead of overwriting the previous one, so one card can hold an image per searched term. Browsing (⬅ ➡) or re‑searching the same query still swaps the last image in place, keeping editor focus and undo history.
  • Graphical Settings Panel: An easy-to-use settings panel to manage your configuration. No more manual file editing!
  • Smart Defaults: Automatically uses the first field of a note type for searching and the last field for placing the image if not configured otherwise.
  • Search on Selection: Simply highlight any text in the editor and use the search button or right-click context menu to search for an image.
  • Toolbar Integration: Adds 🖼, ⬅, and ➡ buttons directly to the Anki editor toolbar for a fast workflow.
  • Right-Click Context Menu: Right-click on highlighted text to instantly start an image search.
  • Support Tab with Supporter Opt-out: A dedicated Support tab with Ko-fi, UPI, BTC, and ETH addresses, and an "I have supported this addon" checkbox that hides the automatic post-update welcome.
  • Update Welcome: When the add-on is updated, the Support tab opens automatically the next time you open the settings dialog (no startup cost — the check runs only on dialog open). The opt-out checkbox in the Support tab turns this off.
  • Built-in Log Viewer: The Logs tab shows a rotating log file (512 KiB × 3). Lazy-loaded — nothing is read from disk until you click Refresh. A "Log debug" checkbox toggles maximum verbosity on/off without picking a level from the dropdown; the "All" level captures even sub-DEBUG noise from third-party libraries.
  • Log Health Scan: A "Check log file" button on the Logs tab scans the log for known error patterns (tracebacks, timeouts, permission errors, provider give-ups, etc.) and shows a per-category summary so you can spot issues without re-reading the whole file.
  • Non-Modal Settings: The settings dialog is non-modal — you can keep editing cards in Anki while it's open. A second click on the menu re-focuses the existing window instead of stacking a new one.

Usage

1. Configuration

  • Open Tools → Image Search v3 Settings, select a note type on the left, then configure Query Fields, the Image Field, and Image Placement on the right, and click Save.
  • Under Network, pick the provider; for Google, paste your API key and CSE ID (cx), and adjust timeout, retries, and backoff to suit your network.
  • Defaults ship with an empty key/cx and provider set to Yandex, and sane network values for reliable startup on new installs.
Screenshot_20251104_012633 Screenshot_20251104_020651

2. Searching for Images

There are three ways to search for an image in the card editor:

  1. Using the Toolbar Button: Click the 🖼 button on the editor toolbar.
    • If you have text highlighted anywhere in the editor, that text will be used for the search.
    • If no text is highlighted, the content of your configured Query Field(s) will be used.
  2. Using the Right-Click Menu: Highlight the text you want to search for, right-click it, and select "Search image for: '...'" from the context menu.
  3. Browsing Results: Use the ⬅ and ➡ buttons to browse through other image results for the last query that was performed from the query field(s).
Screenshot_20251031_152224 Screenshot_20251031_152301 Screenshot_20251031_152339

3. Support tab & post-update welcome

  • Open Tools → Image Search v3 Settings → Support to find Ko-fi, UPI, BTC, and ETH addresses.
  • If you have supported the add-on, tick "I have supported this addon" at the top of the Support tab. The add-on records your preference in meta.json (not in your committed config) and stops opening the Support tab on every update.
  • The post-update welcome only runs when you open the settings dialog (not at Anki startup), so it has zero effect on startup time.
  • To disable the welcome entirely (for example in a managed environment), set auto_show_support_on_update to false in your add-on config (the Logs tab is the easiest place to edit JSON via Tools → Image Search v3 Settings → Logs; or use Tools → Add-ons → Image Search v3 → Config).

Provider notes

  • Yandex: no‑auth, undocumented JSON endpoint used by the front‑end; works well but may change, be geo‑restricted, or rate‑limited without prior notice.
  • Bing: no‑auth, undocumented /images/async endpoint that returns ~30–50 results per query. No JavaScript rendering required; we extract the murl field from the response HTML.
  • Brave: first-party Image Search API. Requires a paid plan and the X-Subscription-Token header; we use the properties.url (original) or thumbnail.src (Brave-proxied) fields from the response. Up to 200 results per request.
  • DuckDuckGo: hidden i.js endpoint (no API key); works best‑effort and may change, rate‑limit, or block without notice. A Linux Chrome User-Agent is required because DDG returns 403 for some Windows UAs. Uses the configured global fallback chain if no results.
  • Google: official Custom Search JSON API with searchType=image; requires both API key and CSE (Google Search Engine) ID (cx) and enforces quotas and billing on your account.
  • Routing: the selected provider is tried first, then the configured global fallback providers in order, preserving the editing flow when any provider returns no results or errors.

If you don't know how to get the API please read this: google custom-search

Brave API key: https://brave.com/search/api/

Developer documentation lives under doc/ (architecture, modules, JSON keys, errors, build, security, testing).

Troubleshooting

Context Menu Item Not Appearing

If the "Search image for..." option does not appear when you right-click on selected text, it might be due to a conflict with another add-on that also modifies the context menu. A common conflict is with add-ons that provide image editing or other right-click functionalities in the editor.

You can diagnose this by temporarily disabling other editor-related add-ons (like "Image Editor") via Tools -> Add-ons, restarting Anki, and checking if the menu item appears.

Support

If you find this add-on useful, please consider supporting its development:

ko-fi

Changelog

The full release history is in CHANGELOG.md (newest first, grouped by version, with links to the corresponding GitHub releases). Contributor and maintainer documentation (architecture, modules, JSON keys, errors, build, security, testing) lives under doc/.

License

This add-on is a modification of the work of original authors. Credit goes to the creators of Anki Image Search v2 and Image Search.

Toolbar buttons use emoji labels to avoid bundling image assets.

This project is licensed under the GPLv2.