About flashdeck
flashdeck is an MCP server and CLI for studying and managing Anki flashcards through a local Anki instance with AnkiConnect. It supports review actions as well as deck, note, tag, and collection-statistics management.
Key features
- 01Review due cards and rate reviews
- 02List and create decks
- 03Create, search, update, and delete notes
- 04Manage note tags and types
- 05View deck and collection statistics
- 06Sync with AnkiWeb
Use cases
- 01Review cards due in a selected deck
- 02Add or batch-create notes in Anki
- 03Find notes and update their fields
- 04Inspect study and collection statistics
MCP support
The MCP server exposes tools for card reviews, decks, notes, tags, note types, synchronization, and statistics. It requires Anki running with the AnkiConnect add-on.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
20Saved tool list · updated 1w ago
This tool takes no parameters.
Details
From the project READMEflashdeck
flashdeck lets an agent study and manage Anki flashcards through a local Anki instance running AnkiConnect. It covers the review loop (due cards, presenting a card, rating it), decks, notes, tags, note types, and collection statistics — a labelled core subset of 20 tools from the 53 documented in the original Anki MCP spec.
Anki must be running with the AnkiConnect add-on installed. By
default flashdeck talks to http://127.0.0.1:8765; set
ANKICONNECT_URL to point elsewhere.
Install
npx @reneturcios/flashdeck check
Or run the container image:
docker run -i ghcr.io/cloudcorprecords/flashdeck
The container defaults to the MCP server. Pass CLI arguments to use it as a CLI instead. Note the container still needs network access to the host's AnkiConnect endpoint.
CLI
flashdeck check # probe AnkiConnect (deck list on success)
flashdeck decks # list decks
flashdeck due [deck] # due cards, optionally for one deck
flashdeck stats [deck] # deck stats, or collection stats with no deck
flashdeck tags # all tags
flashdeck models # all note types
flashdeck sync # sync with AnkiWeb
flashdeck --version
Most commands accept --json. Exit codes: 0 success, 1 error
(the message names the AnkiConnect or validation problem), 2
usage error.
MCP server
npx -y @reneturcios/flashdeck mcp
MCP client configuration:
{
"mcpServers": {
"flashdeck": {
"command": "npx",
"args": ["-y", "@reneturcios/flashdeck", "mcp"]
}
}
}
Tools
| Tool | Input | What it does |
|---|---|---|
sync | — | Sync with AnkiWeb. |
get_due_cards | deck?, include_answer? | Due cards, optional deck filter; answers omitted unless requested. |
get_cards | deck?, state?, include_answer? | Cards filtered by state (due/new/learning/suspended/buried) and deck. |
present_card | card_id | Show a card's question/front. |
rate_card | card_id, ease (1-4) | Rate Again/Hard/Good/Easy and schedule next review. |
listDecks | include_stats? | List decks, optional per-deck stats. |
deckStats | deck | Statistics for one deck. |
createDeck | deck | Create a deck (Parent::Child, max 2 levels). |
addNote | deck_name, model_name, fields, tags? | Create one note. |
addNotes | deck_name, model_name, notes | Batch-create up to 100 notes. |
findNotes | query | Search notes with Anki query syntax. |
notesInfo | notes | Details for notes. |
updateNoteFields | note_id, fields | Update a note's fields. |
deleteNotes | notes, confirm? | Delete notes; refused without confirm: true. |
getTags | — | All tags. |
addTags | notes, tags | Add tags to notes. |
removeTags | notes, tags | Remove tags from notes. |
modelNames | — | List note types. |
modelFieldNames | model_name | Field names for a note type. |
collection_stats | — | Aggregated collection statistics. |
Deck statistics are computed locally from per-state card searches; no statistics endpoint is assumed. When Anki is not running, tools return an error naming the failed AnkiConnect request rather than an empty result.
Development
npm install
npm run build # tsc -> dist/
npm test # node test runner over test/*.test.ts, fake fetch only, no live Anki
npm run dev # run the CLI from source via tsx
License
MIT — see LICENSE.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.