Best for
- Individual learners already using Anki and AnkiConnect.
- Users who want AI-guided review, explanations, and dynamic card creation.
- Users connecting local Anki to Claude Desktop, Cursor, Cline, or Zed.
Based on the supplied repository, manifest, and README, this is an independent MIT-licensed Anki integration that exposes 42 MCP tools through AnkiConnect. The manifest lists version 0.18.4, while the README's current-status section states 0.22.0, so the version information is inconsistent.
Strongest · Documentation 15/20 Weakest · Reliability 9/20
Static review · not runListed 2026-07-29
Read the FMRS scoring method →Runtime:Node.js 22.12.0+
ANKI_CONNECT_API_KEY
Set only if you configured an API key in AnkiConnect.
TUNNEL_AUTH_CLIENT_ID
OAuth client ID for the device flow; only needed for self-hosted tunnel/auth services.
ANKI_CONNECT_URL
URL of the AnkiConnect plugin, default http://localhost:8765; usually no change needed.
ANKI_CONNECT_API_VERSION
AnkiConnect API version, defaults to 6.
ANKI_CONNECT_TIMEOUT
Request timeout in ms, default 5000.
READ_ONLY
Set to true or 1 to block all write operations.
ALLOWED_HOSTS
Extra Host header values accepted in HTTP mode; required when binding to a LAN/public address.
ALLOWED_ORIGINS
Comma-separated allowlist of browser Origin/Referer patterns for HTTP mode; wildcards supported.
TUNNEL_SERVER_URL
Tunnel service WebSocket URL for tunnel mode, default wss://tunnel.ankimcp.ai.
MEDIA_ALLOWED_TYPES
Extra MIME types allowed for file path imports, e.g. application/pdf.
MEDIA_IMPORT_DIR
Restricts file path imports to this directory.
MEDIA_ALLOWED_HOSTS
Private network hosts allowed for URL imports.
PORT
HTTP server port, default 3000.
HOST
HTTP bind address, default 127.0.0.1.
LOG_LEVEL
Logging level, default info.
Install Anki and the AnkiConnect plugin, and confirm that AnkiConnect uses http://localhost:8765. Install Node.js 22.12.0 or newer. Run the server with npx and add the configuration above to an MCP client that supports STDIO; Claude Desktop can also install the project's MCPB bundle.
{"mcpServers":{"anki-mcp":{"command":"npx","args":["-y","@ankimcp/anki-mcp-server","--stdio"],"env":{"ANKI_CONNECT_URL":"http://localhost:8765"}}}}
Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.
{
"servers": {
"anki-mcp": {
"command": "npx",
"args": [
"-y",
"@ankimcp/anki-mcp-server",
"--stdio"
],
"env": {
"ANKI_CONNECT_URL": "http://localhost:8765"
}
}
}
}
Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).
claude mcp add anki-mcp -e ANKI_CONNECT_URL=http://localhost:8765 -- npx -y @ankimcp/anki-mcp-server --stdio
Run it in a terminal; replace any <…> placeholders with your own values first.
After starting the server, tools like sync, listDecks, and get_due_cards should appear in your MCP client's tool list; asking 'List my Anki decks' and getting real deck data back confirms the AnkiConnect connection works.
Once connected, you can ask your AI assistant things like:
No matching tools
Listed from the project's documentation, not tested by this site.
Anki MCP Server connects AI assistants to Anki flashcards through AnkiConnect. It supports reviews, decks, notes, media, tags, models, templates, statistics, and selected Anki desktop GUI workflows. The project is in active beta development, so APIs and features may change.
Manage your Hevy workout data from AI assistants via MCP.
Firecrawl's official MCP server for web search, scraping, and structured extraction for AI agents
One binary. One grammar. Evidence from the biomedical sources you already trust.
Encrypted, fully offline long-term memory for AI agents
Source revision e0e347f7f399 Data synced 2026-10-19 Read the FMRS scoring method