Skip to main content
Glama
README.md
## TranslationX MCP Server (Python)
### 搭建Python虚拟环境
我们推荐通过`uv`构建虚拟环境来运行MCP server,关于`uv你可以在[这里](https://docs.astral.sh/uv/getting-started/features/)找到一些说明。

按照[官方流程](https://modelcontextprotocol.io/quickstart/server),你会安装`Python`包管理工具`uv`。除此之外,你也可以尝试其他方法(如`Anaconda`)来创建你的`Python`虚拟环境。

通过`uv`添加`mcp`依赖

```bash
uv add "mcp[cli]"
```

验证mcp依赖是否安装成功,执行如下命令
```bash
uv run mcp
```

当出现下图时代表安装成功

![](../../../img/uv_install_success.png)

通过`uv`安装`python`,最低版本要求为3.13

```bash
uv python install 3.13
```

### 获取 MCP Server
前往TranslationX Mcp Server 官方[开源仓库](https://github.com/Deep-Intelligent-Pharma/Translationx-mcp-server/)下载

### 配置本地项目
通过`uv`创建一个项目

```bash
uv init mcp_server_translation
```

将`src`文件夹拷贝到该目录下,通过如下命令测试mcp server是否正常运行

```bash
uv run --with mcp[cli] mcp run {YOUR_PATH}/src/main.py
# 如果是mac,需要加转义符
uv run --with mcp\[cli\] mcp run {YOUR_PATH}/src/main.py
```

如果没有报错则MCP Server启动成功

### 在Cursor中使用

打开`Cursor`配置,在MCP中添加MCP Server

![](img/cursor_setting.png)

在文件中添加如下内容后保存, TRANSLATIONX_TOKEN 可以在[X-DOC](https://x-doc.ai/)中获取

```json
{
  "mcpServers": {
    "translationx": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "mcp[cli]",
        "mcp",
        "run",
        "{YOUR_PATH}/src/main.py"
      ],
      "env": {
        "token": "<YOUR_TRANSLATIONX_TOKEN>"
      }
    }
  }
}
```

回到配置,此时TranslationX MCP Server已经启用

![](img/cursor_run_mcp_success.png)

### 测试

创建项目,并创建文件夹,然后文件到文件夹中,并提交翻译:

![](img/cursor_test_1.png)
![](img/cursor_test_1_1.png)

![](/img/cursor_test_2.png)

TDQS

C2.6/5.0

Scored across 22 tools

Disambiguation4/5

Most tools have distinct purposes such as managing memory libraries, term libraries, projects, and files. However, there is potential confusion between 'delete_file' and 'delete_project_file', and between 'file_add' and 'upload_project_file', as their descriptions are brief and overlapping.

Naming Consistency3/5

Tool names mix verb_noun patterns (e.g., 'create_memory_lib') with noun_verb patterns (e.g., 'file_add'), and some use different verbs for similar actions (e.g., 'edit_project' vs 'delete_project'). The naming is not fully consistent.

Tool Count5/5

With 22 tools, the server covers core functionalities for a translation management system, including project, file, memory, and term library operations. The count is well-scoped and reasonable for the domain.

Completeness3/5

The tool set covers basic CRUD for projects and folders, file upload/download, and translation submission. However, it lacks update/delete for library entries, retrieval of single project details, and translation status checking, leaving notable gaps in the workflow.

Maintenance

ActivityInactive
ResponsivenessNo issues