A browser extension that aims to parse most Japanese text in the browser using JPDB. Support for mining into Anki decks is in preparation and will be added in the future.
Thanks to Max and the JPDB Discord for making the original extension and fixing some issues along the way!
Sadly, Manifest V3 came along and the original is no longer working.
Some web apps and sites require special attention to work properly; therefore, they parse automatically on certain triggers. These can be disabled or extended in the settings.
| App | URLs |
|---|---|
| ッツ Reader | reader.ttsu.app, ttu-ebook.web.app |
| Texthooker pages | anacreondjt texthooker, learnjapanese.moe texthooker, exSTATic tracker, renji-xd texthooker |
| Mokuro | reader.mokuro.app |
| Mokuro (Legacy) | Mokuro: IMPORTANT: File path must contain mokuro, and file name must end in .html |
| Readwok | app.readwok.com |
| Satori Reader | satorireader.com |
| NHK News | nhk.or.jp |
| Wikipedia | ja.wikipedia.org |
| Bunpro | bunpro.jp (Experimental) |
| asbplayer | github.com/killergerbah/asbplayer |
| YouTube comments | youtube.com |
| Custom Dictionaries | CDM by Nakura Nakamoto |
| LunaTranslator | Luna Translator with WebView2 enabled |
*-chromium.zip file from the releases page.chrome://extensions/.Developer mode (top right).Load unpacked button (top left).manifest.json inside the folder.Firefox Browser Extension, or install it via Firefox for Android or
-firefox.xpi file from the releases page.about:debugging.This Firefox, then Load Temporary Add-on.Please note that if you use the manual setup, the extension will be unloaded the next time you open Firefox. Your settings will be preserved. This is only for testing and debugging!
I am currently not able to test or verify the Edge version of the extension, but you can follow the instructions on the Kiwi browser repository as a starting point.
The extension ID is gpomfklkdbhmpagecjpnlhlffdlhcbnb.
Upon installation, the settings page will open. You can also find it by clicking on the reader icon (読) in the browser menu bar. It might be hidden behind the extension overflow menu, which looks like a little puzzle piece (🧩).
Here you will need to enter your JPDB API key. It can be found at the very bottom of the JPDB settings page.
You can also change various hotkeys.
You can use the reader on any website. Just select some text, right-click, and choose the "Parse ... with JPDB" option. Alternatively, use the shortcut (Alt+P) or the extension menu at the top right corner.
Words will be colored according to their state (known, new, etc.). Hover over words while holding Shift to see their meaning, and to mine or review them.
Parsing can be enabled or disabled per integration. You can also add custom URLs to automatically parse in the settings.
Additionally, you can add complete meta definitions in JSON format - refer to the typings or see the docs
Customization is currently done with custom CSS, as it is the simplest way to offer a flexible framework. See the docs on it here.
Node version 22.x is used!
You can run the following commands to build the extension locally:
npm install
npm run build
The resulting files will be located in the anki-jpdb.reader/ folder.
For development, you can also run the build in watch mode:
npm install
npm run watch
This will continuously rebuild the source code as it changes, and place the output in the anki-jpdb.reader/ folder.
It can be loaded as an unpacked extension from there.
Please remember to wait until building is done, and reload the extension on the "Manage Extensions" page before testing your changes.
Reloading is not required for the browser popup (top right) or the settings page.
Also, please look at the Contributing section if you plan on contributing your changes.
Issues with feedback or ideas for new features are very welcome. You can also message me on the JPDB and Refold Japanese Discord servers (@chinokusari).
The following commands may be of interest to you:
* npm run lint: Checks your code for formatting issues, linter warnings, and type errors. The CI also runs this, so your pull request will only be accepted if it passes. You can use eslint ignore comments if you get false positives, but leave a comment explaining why you think the error is false and safe to ignore.
* npm run lint:fix: Reformats your code, as well as fixing any fixable lint issues. Note: if your editor has a prettier plugin, installing that and turning on "format on save" will be more convenient.
* npm run build: Compiles the code for Chrome, putting the compiled code into anki-jpdb.reader/.
* npm run build [target]: Builds for a specific target.
* npm run watch: Automatically recompiles the code for Chrome when it changes, putting the output into anki-jpdb.reader/. Using this is recommended during development.
* npm run watch [target]: Recompiles for the specified target.
* npm run pack: Builds, then packs the extension for all targets. The archives are placed inside packages/.
* npm run pack [target, [target...]]: Builds for specified targets.
Currently, chrome|chromium and firefox are supported.
This project uses Font Awesome 4, which is licensed under the MIT License. Please ensure compliance with its license when using or distributing this project.
$ claude mcp add anki-jpdb.reader \
-- python -m otcore.mcp_server <graph>