Min Anki	Max Anki	Updated
2.1.1	+	2026-04-13

Disclaimer

Due to filesize issues the deck (old version) with audiofiles can be found on https://drive.google.com/file/d/1P3xBSHHB26fl5I7XyeRaTjwXtzGfPcpA/view?usp=sharing

The light version without the audiofiles is here

You cannot rename the deck otherwise the crawler addon will not work (so rename it to "grammar" after you downloaded it)

JLPT Grammar Graph Database

This is a relational grammar database that provides an overview of different Japanese grammar patterns and their connections. These connections can be visualized via add-ons.

The note type is designed so that only the back is visible. It functions as a database and intermediary tool during the sentence mining process; it is not intended for traditional flashcard reviewing, but rather for looking up reference information.

🔷 Add-ons specifically for this grammar note

These optional add-ons enhance the functionality of the note type but are not strictly required:

Manual Linker Add-on: AnkiWeb ID 744725736 – Allows establishing connections manually by selecting a group of cards to mutually link them to each other. (Based on the Notelink Add-on installed during setup).
Relation Degree Graph & Crawler: AnkiWeb ID 1321136162 – The engine used for visualizing the grammar network and crawling connections.

Note Visuals and Functionality Demo

https://github.com/user-attachments/assets/0e4ca214-63b5-4fa3-8457-0beb05aa9284

Crawler Addon Demo

With the Crawler Addon you can create clusters and with the help of the NoteLinker addon (see setup) you can also visualize the connections in a anki native overview

The " みたい" cluster of degree 2 and 3:

Video Link for degree 2: https://github.com/user-attachments/assets/01ecc6cf-03b4-4220-b0d1-26857b3ef7c2

Video Link for degree 3: https://github.com/user-attachments/assets/2b9c75aa-e750-413a-bc37-0b184a35c306

Big Graph Demo

Big Graph Demo Video click link:

https://github.com/user-attachments/assets/635b9d71-c994-43cc-bddb-f8bd3170f4fc

Field Descriptions

The note structure consists of the following fields:

Field Name	Description
Level And Grammar Point	Contains the JLPT level prefix, followed by a representative Japanese string identifying the grammar, and a condensed meaning description.
Link	A URL linking to the original JLPT Sensei website entry. This field is empty for manually created cards.
Connected Grammar Points from jlptsensei	Contains links to syntactically or semantically similar grammar points. These were copied from JLPT Sensei during the initial setup via the Notelinker add-on. Links can also be added manually using the Manual Linker add-on.
Notes	Personal notes and custom reminders related to the grammar structure.
construction	The grammar connection rules, copied from JLPT Sensei. This field serves as the structural baseline for generating the corresponding regex patterns. For manual cards, this content can be generated with AI, keeping the underlying HTML structure intact to preserve the layout.
examplesentences	Sample sentences copied from JLPT Sensei. For manual cards, this can contain personal examples or sentences encountered during reading.
title audio	TTS audio file reading out the text in the Level And Grammar Point field.
construction audio	TTS audio file reading out the text in the construction field.
examplesentence audio	TTS audio file reading out the text in the examplesentences field.
regexpattern	Regular expression designed to match the specific grammar pattern. It is read by the grammarhub browser extension to detect and match grammar structures on web pages.
mined sentence	Contains the context sentences collected via grammarhub. It is populated automatically with bidirectional links by the grammarminer add-on.

Usecases

1. The Relation Crawler & Filtered Decks

This workflow utilizes the Relation Degree Graph & Crawler add-on to build specific study sets.

Select Cards: Highlight base cards in the Anki Browser.
Start the Crawler: Go to Edit -> Tag _Process Degree Relations.
Configure Cluster: Choose your search depth (Degree).
Create Study Session: Use the add-on's built-in feature to automatically generate a Filtered Deck for the identified grammar cluster.
Interactive Learning: Use the direct links on the back of each card to jump to related points.

2. Browser Mining with GrammarHub & GrammarMiner

This workflow automates the extraction of contextual sentences from web pages and links them directly to your base database.

Pattern Matching: The grammarhub browser add-on reads the regexpattern field from your database to detect, match, and display grammar points directly in the browser while reading.
Sentence Collection: When you mine a sentence, grammarhub stores the contextual data and generates two TSV files.
Database Update: The grammarminer Anki add-on processes these TSV files. It automatically creates new, individual simplemining grammar cards for your active review and updates the underlying base cards with the mined sentences, establishing bidirectional links between the base card and the mining card.

Merging Audio from the Legacy Deck Version

To import the audio files from the old deck version into the updated format, use the following local media directory method:

Download the legacy deck version containing the audio files from the provided Google Drive link.
Import the legacy deck into Anki. This automatically extracts and stores all audio files into your local collection.media folder.
Delete the legacy deck from Anki. The audio files will remain in your local media folder.
The updated cards will automatically recognize and play the audio files because the file naming convention has not changed.

JLPT Database graph relation crawler

AnkiWeb

Maintenance

GitHub

Will it work on my Anki?

Similar addons

README