kitagry/anki-mcp
If you are the rightful owner of anki-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
Anki MCP Server is a Model Context Protocol server that facilitates interaction between LLM applications and Anki flashcards using AnkiConnect.
Anki MCP Server
A Model Context Protocol (MCP) server that enables LLM applications to interact with Anki flashcards through AnkiConnect.
Features
- Deck Management: Get list of all available decks
- Note Creation: Add new flashcards to any deck
- Note Search: Find notes using Anki's search syntax
Prerequisites
- Anki must be running
- AnkiConnect plugin must be installed in Anki
- Install from Anki: Tools > Add-ons > Browse & Install
- Code:
2055492159
- Restart Anki after installation
Installation
git clone https://github.com/kitagry/anki-mcp
cd anki-mcp
go build
Usage
Run the MCP server:
./anki-mcp
The server communicates via stdio and provides the following tools:
Available Tools
deck_names
Get all deck names from Anki.
Parameters:
anki_url
(optional): AnkiConnect URL (defaults to http://localhost:8765)
add_note
Add a new flashcard to Anki.
Parameters:
anki_url
(optional): AnkiConnect URL (defaults to http://localhost:8765)deck_name
(required): Name of the deck to add the note tomodel_name
(optional): Note model name (defaults to 'Basic')front
(required): Front content of the noteback
(required): Back content of the notetags
(optional): Comma-separated tags
add_notes
Add multiple flashcards to Anki in a single operation.
Parameters:
anki_url
(optional): AnkiConnect URL (defaults to http://localhost:8765)deck_name
(required): Name of the deck to add the notes tomodel_name
(optional): Note model name (defaults to 'Basic')notes
(required): JSON array of notes with 'front', 'back', and optional 'tags' fields
Example notes format:
[
{
"front": "What is the capital of Japan?",
"back": "Tokyo",
"tags": ["geography", "capitals"]
},
{
"front": "What is 2+2?",
"back": "4",
"tags": ["math"]
}
]
find_notes
Find notes in Anki using a search query.
Parameters:
anki_url
(optional): AnkiConnect URL (defaults to http://localhost:8765)query
(required): Search query (e.g., 'deck:Default', 'tag:important')
Configuration
By default, the server connects to AnkiConnect at http://localhost:8765
. You can specify a different URL using the anki_url
parameter in each tool call.
Dependencies
- mark3labs/mcp-go - Go implementation of Model Context Protocol
- AnkiConnect - Anki plugin for HTTP API access
License
MIT