Skip to main content
Glama
oofengoo

MCP-Serve 多语言翻译服务

by oofengoo
README.md
# MCP-Serve 多语言翻译服务

## 项目介绍

MCP(Model Context Protocol)服务是一种用于扩展 AI 模型能力的协议实现,允许 AI 模型与外部服务进行交互,从而增强其功能。本项目是一个基于 MCP 协议的自定义服务,专门用于处理多语言翻译和资源管理。

## 功能特点

本项目提供两个主要功能:

### 1. 多语言翻译 (translationI18n)

- 自动识别并提取文本中被 `##` 包裹的内容(如 `##创建人##`)
- 将提取的内容发送到指定的 API 进行多语言资源注册
- 支持自动生成资源别名(ResourceAlias)

### 2. 多语言资源复制 (copyI18n)

- 从 API 下载最新的多语言资源文件
- 自动将资源文件复制到项目的语言目录中(`src/lang/source/`)

## 在 VSCode 中使用

### 1. 构建项目

首先,克隆项目并安装依赖:

```bash
git clone https://github.com/oofengoo/mcp-serve-learn
cd mcp-serve
npm install
```

然后构建项目:

```bash
npm run build
```

### 2. 配置 VSCode

将`mcp-servers.json`文件中的配置复制到 VSCode 设置中:

1. 打开 VSCode 设置(文件 > 首选项 > 设置)
2. 点击"在 settings.json 中编辑"
3. 将以下内容添加到 settings.json 中:

```json
{
  "mcp": {
    "servers": {
      "mcp-server": {
        "command": "node",
        "args": [
          "D:\\Users\\Administrator\\Documents\\mcp-serve\\dist\\index.js"
        ],
        "env": {
          "Token": "",
          "URL": "",
          "PORT": ""
        }
      }
    }
  }
}
```


## 环境变量说明

- `Token`: API 认证令牌(Cookie 值)
- `URL`: API 服务器地址
- `PORT`: 服务端口

## 开发说明

本项目使用 TypeScript 开发,主要文件结构:

- `src/index.ts`: MCP 服务器实现,包含工具定义
- `src/api.ts`: API 接口封装
- `src/request.ts`: HTTP 请求工具

TDQS

D1.2/5.0

Scored across 2 tools

Disambiguation1/5

The two tools have overlapping purposes with no descriptions to clarify differences. 'copyI18n' and 'translationI18n' both appear to handle i18n (internationalization) tasks, making them indistinguishable. An agent cannot determine when to use one versus the other, leading to high misselection risk.

Naming Consistency3/5

The naming shows mixed conventions: 'copyI18n' uses camelCase while 'translationI18n' uses a similar style but with a different prefix. Both include 'I18n', providing some thematic consistency, but the lack of a clear verb_noun pattern and the variation in prefixes ('copy' vs. 'translation') reduces predictability.

Tool Count2/5

With only 2 tools for a '多语言翻译服务' (multilingual translation service), the count feels too thin. A translation service typically requires more operations, such as translate_text, detect_language, or manage_translations. This minimal set limits functionality and suggests an incomplete implementation.

Completeness1/5

The tool surface is severely incomplete for a translation service. There are no tools for core operations like translating text, detecting languages, or managing translation keys. The two vague tools do not cover the domain's essential workflows, leaving significant gaps that will cause agent failures.

Maintenance

ActivityInactive
ResponsivenessNo issues