Skip to main content
Glama

Eudic MCP Server

欧路词典(Eudic)生词本 API 的 Model Context Protocol 封装。

支持 Claude Desktop、Cursor、Kimi Code 等兼容 MCP 的客户端直接管理 Eudic 生词本分组与单词。

功能

  • 列出/创建/重命名/删除生词本分组

  • 列出/添加/删除生词本中的单词

  • 批量导入单词

  • 查询已掌握单词

  • 管理单词笔记(增删改查)

  • 获取用户例句列表

Related MCP server: mcp-server-ticktick

安装

从 PyPI 安装(暂未发布)

pip install eudic-mcp

从源码安装

git clone https://github.com/safeblood/eudic-mcp.git
cd eudic-mcp
pip install -e .

不安装,直接运行源码

如果你不想安装到系统 Python,可以设置 PYTHONPATH 指向项目目录:

export PYTHONPATH="/path/to/eudic-mcp"
python -m eudic_mcp.server

配置

需要欧路词典 OpenAPI Token。获取方式见 Eudic OpenAPI 文档

注意:Token 只填 NIS 后面的字符串部分,不要带 NIS 前缀。本客户端会自动拼上 NIS

# 错误
export EUDIC_API_TOKEN="NIS xxxx"

# 正确
export EUDIC_API_TOKEN="xxxx"

Windows PowerShell:

$env:EUDIC_API_TOKEN="xxxx"

在 Kimi Code 中使用

编辑 Kimi Code 的 MCP 配置文件:

  • Windows: C:\Users\<用户名>\.kimi-code\mcp.json

  • macOS/Linux: ~/.kimi-code/mcp.json

添加或替换 eudic server:

{
  "mcpServers": {
    "eudic": {
      "command": "python",
      "args": [
        "-m",
        "eudic_mcp.server"
      ],
      "env": {
        "PYTHONPATH": "C:/Users/83773/Downloads/pte/eudic-mcp",
        "EUDIC_API_TOKEN": "你的token"
      }
    }
  }
}

修改后重启 Kimi Code 生效。

在 Claude Desktop 中使用

编辑 claude_desktop_config.json

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

  • Windows: %APPDATA%\Claude\claude_desktop_config.json

已安装包

{
  "mcpServers": {
    "eudic": {
      "command": "eudic-mcp",
      "env": {
        "EUDIC_API_TOKEN": "你的token"
      }
    }
  }
}

直接运行源码

{
  "mcpServers": {
    "eudic": {
      "command": "python",
      "args": [
        "-m",
        "eudic_mcp.server"
      ],
      "env": {
        "PYTHONPATH": "C:/Users/83773/Downloads/pte/eudic-mcp",
        "EUDIC_API_TOKEN": "你的token"
      }
    }
  }
}

可用工具

工具名

说明

eudic_list_categories

列出所有生词本分组

eudic_create_category

创建分组

eudic_rename_category

重命名分组

eudic_delete_category

删除分组

eudic_list_words

列出分组单词

eudic_add_word

添加单个单词

eudic_add_words_bulk

批量导入单词

eudic_delete_words

批量删除单词

eudic_get_word

查询单个单词

eudic_list_mastered_words

查询已掌握单词

eudic_list_notes

获取所有单词笔记

eudic_get_note

获取某个单词的笔记

eudic_add_note

添加或更新单词笔记

eudic_delete_note

删除单词笔记

eudic_list_sentences

获取用户例句列表

工具调用示例

创建生词本分组

{
  "name": "wfd",
  "language": "en"
}

返回示例:

{
  "data": {
    "id": "132314173819830130",
    "language": "en",
    "name": "wfd"
  },
  "message": ""
}

批量导入单词

{
  "words": ["hypothesis", "methodology", "empirical"],
  "category_id": "132314173819830130",
  "language": "en"
}

添加单个单词(带语境)

{
  "word": "hypothesis",
  "category_ids": ["132314173819830130"],
  "star": 3,
  "context_line": "This hypothesis needs more evidence.",
  "language": "en"
}

列出分组单词

{
  "category_id": "132314173819830130",
  "page": 0,
  "page_size": 100,
  "language": "en"
}

添加/更新单词笔记

{
  "word": "hypothesis",
  "note": "假设;假说。注意与 assumption 区分。",
  "language": "en"
}

获取单词笔记

{
  "word": "hypothesis",
  "language": "en"
}

删除单词笔记

{
  "word": "hypothesis",
  "language": "en"
}

列出用户例句

{
  "language": "en",
  "page": 0,
  "page_size": 100
}

作为 Python 库使用

from eudic_mcp import create_category, add_words_bulk

cat = create_category("wfd")
cat_id = cat["data"]["id"]

add_words_bulk(["hypothesis", "methodology"], category_id=cat_id)

开发

cd eudic-mcp
pip install -e .

# 手动启动 server
EUDIC_API_TOKEN="你的token" eudic-mcp

依赖

  • Python >= 3.10

  • mcp>=1.0.0

  • requests>=2.28.0

License

MIT

Related MCP Connectors

Related MCP Servers

  • F
    license
    Not graded
    quality
    C
    maintenance
    MCP server for Evernote China (印象笔记), enabling Claude to list notebooks and notes, search, read, create, and update notes via natural language.
    1
    -
  • A
    license
    A
    quality
    B
    maintenance
    Enables reading and writing to your local Anki collection through AnkiConnect, allowing creation, update, and search of notes and decks from MCP clients like Claude Desktop.
    11
    MIT