WaniKani MCP Server

what is this?

this is an MCP (Model Context Protocol) server that provides Claude with access to your WaniKani data for japanese kanji and vocabulary learning.

note: this is an unofficial implementation and is not affiliated with WaniKani or Tofugu.

available tools

⏺ get_subject_by_id Get subject by its subject id. A subject is a i64 identifier which can be used for kanji, radicals or vocabulary

⏺ find_new_kanji_from_sentence Find new kanji and their wanikani levels from a given sentence that the user has not learned yet

⏺ get_level_info Get level information

⏺ find_new_vocabulary Find new vocabulary and their wanikani levels from a list of words/phrases that the user has not learned yet

⏺ get_kanji_by_character Get multiple kanji subjects by its characters. Returns the level, onyomi, kunyomi, reading mnemonic and meaning mnemonics, and radicals that compose the kanji

⏺ get_vocabulary_by_phrase Get vocabulary subject by its phrase. Returns the readings, meaning mnemonics and example sentences, and kanji subject IDs that compose the vocabulary

setup instructions

option 1: use the publicly hosted server (easiest)

{
  "mcpServers": {
    "wanikani": {
      "url": "https://wanikani-mcp-666563326154.us-west1.run.app/mcp/sse",
      "transport": {
        "type": "sse"
      }
    }
  }
}

option 2: run your own server (recommended)

{
  "mcpServers": {
    "wanikani": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/srevinsaju/wanikani-mcp:main",
        "--mode",
        "stdio",
        "--api-key",
        "YOUR_WANIKANI_API_KEY"
      ]
    }
  }
}

requires docker. get your api key from wanikani settings