What does the server use to store data?
It interacts with a local Anki collection file (`collection.anki2`). The path can be set via the `ANKI_COLLECTION_PATH` environment variable; otherwise it uses Anki’s default location.
@MCP-Mirror
基本信息
Mirror of
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"vionwinnie_jap-vocab-anki-mcp-server": {
"command": "python",
"args": [
"-m",
"anki_mcp.server"
]
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
A Model Context Protocol server that allows Language Models to interact with Anki decks programmatically, with special support for Japanese language learning. It enables listing decks, viewing and adding cards, reviewing with spaced repetition, and importing Japanese vocabulary with readings, meanings, and sample sentences.
Clone the repository, ensure Anki is not running, optionally set the ANKI_COLLECTION_PATH environment variable, then run python -m anki_mcp.server. Use the provided tools, resources, and prompts to manage Anki decks and study Japanese vocabulary through an MCP-compatible client (e.g., Claude).
It interacts with a local Anki collection file (`collection.anki2`). The path can be set via the `ANKI_COLLECTION_PATH` environment variable; otherwise it uses Anki’s default location.
The server expects a note type named "Japanese (recognition)" with fields: Expression, Meaning, and Reading (furfurigana and sample sentences go in the Reading field).
Yes, basic card management tools (`add_card`, `review_card`, `get_card_history`) work with any deck, but the Japanese-specific tools (import, sentence update) require the "Japanese (recognition)" note type.
The server runs as a standard MCP server using stdio transport (started via `python -m anki_mcp.server`). It does not expose an HTTP or WebSocket endpoint.
Anki must not be running at the same time to avoid database locks. The server only accesses the local Anki collection; it does not sync with AnkiWeb automatically.

Handl speaks the Model Context Protocol. Your whole billing operation — projects, invoices, clients, reports, approvals — as tools any AI assistant can call, inside your autonomy rules.
Hindsight: Agent Memory That Learns
Open-source MCP to use Blender with any LLM

Model Context Protocol Servers

Model Context Protocol Servers
评论