mcp-local-translate
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CACHE_DIR | No | 模型缓存目录。NLLB-200 模型（~600MB）自动下载到此。建议使用绝对路径。 | 系统默认 |
| HF_ENDPOINT | No | HuggingFace 镜像端点。设置后跳过自动镜像检测。 | https://huggingface.co |
| HTTPS_PROXY | No | HTTP 代理地址，用于下载模型（如 http://127.0.0.1:7890）。通过 setGlobalDispatcher 全局生效。 | |
| HF_AUTO_MIRROR | No | 自动镜像检测开关。设为 false 禁用。默认开启，首次下载前自动探测三级镜像链。 | true |
| TRANSLATE_DEVICE | No | 推理设备。直接传给 Transformers.js。目前仅推荐 cpu。 | cpu |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| translateB | Translate text between 200+ languages using the NLLB-200 model, running entirely on your local CPU. No cloud APIs, no data leaves your machine. |
| list_languagesA | List all 200+ languages supported by the NLLB-200 translation model. Optionally filter/search by keyword. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |