Fanbei
Provides document translation using DeepL as a selectable translation engine. Documents (DOCX, PPTX, PDF, XLSX, TXT, or ZIP/7Z/RAR batches) can be quoted per page, uploaded, translated via DeepL, tracked for progress, and downloaded via time-limited links.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Fanbei Translate /tmp/合同.docx into English with GPT-6.1 Sol, bilingual, and send me the download link"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
翻呗文档翻译 · MCP Server
把翻呗的文档翻译能力接进任何支持 MCP 的 Agent: Claude Desktop / Claude Code、Cursor、Cline、Codex,或你自己写的 Agent。
一个端点,七个工具:列引擎 → 报价 → 上传下单 → 查进度 → 下载。
支持格式:DOCX / PPTX / PDF / XLSX / TXT,ZIP / 7Z / RAR 批量包;
.doc/.xls/.ppt上传时自动转换计价:按页扣积分,1 积分/页起(引擎与附加项不同)
引擎:通用模型 / DEEPL / GPT-6 系列 / Claude 5.5 / Gemini / Grok / 豆包 / DeepSeek / GLM / LongCat / 金山·阿里·有道
1. 拿一个访问令牌
curl -X POST https://trans.sunnytranslation.com/trans/api/token \
-d "username=你的账号" -d "password=你的密码"返回里的 token(fb1. 开头,默认 30 天有效)就是全部凭证。
改密码会让旧令牌立即失效——令牌签名的盐是该账号的密码哈希,服务端不存令牌表。
Related MCP server: GroupDocs.Merger MCP Server
2. 配置客户端
Claude Desktop / Claude Code(stdio)
{
"mcpServers": {
"fanbei": {
"command": "uvx",
"args": ["fanbei-mcp"],
"env": { "FANBEI_TOKEN": "fb1.xxxxx" }
}
}
}也支持用账号密码,让服务自己换令牌(首次调用时登录一次,之后缓存到过期):
"env": { "FANBEI_USERNAME": "你的账号", "FANBEI_PASSWORD": "你的密码" }远程接入(streamable-http)
{
"mcpServers": {
"fanbei": {
"type": "http",
"url": "https://mcp.sunnytranslation.com/mcp",
"headers": { "Authorization": "Bearer fb1.xxxxx" }
}
}
}自托管:
fanbei-mcp --transport http --host 127.0.0.1 --port 9010 --json-response3. 工具
工具 | 说明 | 只读 |
| 积分余额、有效期、令牌到期 | ✓ |
| 引擎清单、每页积分、能力边界 | ✓ |
| 按页数估积分,判断余额够不够 | ✓ |
| 上传并下单(可同步等待完成) | ✗ |
| 单任务进度 + 子文件明细 | ✓ |
| 任务列表(在用 / 回收站) | ✓ |
| 10 分钟有效的下载链接,或直接落盘 | ✓ |
典型对话:
把
/tmp/合同.docx用 GPT-6.1 Sol 翻成英文,双语对照,翻完给我下载链接。
Agent 会依次调用 fanbei_engines → fanbei_quote → fanbei_translate_document
→ fanbei_task_status → fanbei_download。
文件怎么传
场景 | 参数 | 说明 |
本地 stdio |
| 读本机文件,最大 300MB |
文件在公网 |
| 服务端拉取,最大 50MB |
远程 / 无路径 |
| 最大 50MB,注意上下文长度 |
4. 注意
扣费发生在下单那一刻。
auto_confirm=false只上传不扣费,但也不会有进度。status=3才是「翻译完成、可下载」;7只是转换完成。失败任务不会自动退款,走客服申诉。
传统网页引擎(金山/阿里/有道)不支持术语表与全文上下文。
下载链接 10 分钟过期,重新调用
fanbei_download换新的。
5. 开发
python -m venv .venv && .venv/Scripts/pip install -e .
FANBEI_TOKEN=fb1.xxx python tests/selftest.py # 全链路自测
FANBEI_TOKEN=fb1.xxx python tests/selftest.py --read-only # 只读自测6. 许可与条款
Available Tools
7 toolsfanbei_account账户余额ARead-onlyInspect
查询当前令牌对应账号的积分余额、有效期与令牌到期时间。
Inputs: no named parameters.
fanbei_download下载译文ARead-onlyInspect
取成品的下载链接(10 分钟有效)。save_to 给本地目录时,stdio 模式下会直接把文件写盘并返回路径。
Inputs: tran_id (required), save_to.
fanbei_engines可选翻译引擎ARead-onlyInspect
列出可用于翻译的引擎(模型档与传统网页引擎)、每页积分与能力边界。code 用于其它工具的 engine 参数。
Inputs: no named parameters.
fanbei_list_tasks任务列表ARead-onlyInspect
分页列出我的翻译任务(默认只看在用的)。scope 可选 active/trash/all。
Inputs: page, limit, scope, keyword.
fanbei_quote翻译报价ARead-onlyInspect
按页数与选项估算积分消耗,并检查余额是否够用。下单前先报个价。
Inputs: pages (required), ocr, engine, bilingual.
fanbei_task_status查询任务进度ARead-onlyInspect
按 tranId 查单个任务的状态、进度百分比与子文件明细。status=3 表示翻译完成、可以下载。
Inputs: tran_id (required).
fanbei_translate_document翻译文档AInspect
上传一个文档并下单翻译。文件来源三选一:file_path(本地路径,stdio 模式用)、file_url(公网直链)、file_base64(≤50MB,远程模式用)。auto_confirm=True 会立刻扣积分并开始翻译;False 只上传返回 tranId。wait_seconds>0 会同步等待到完成或超时(超时仍可用 fanbei_task_status 继续查)。
Inputs: ocr, engine, file_url, filename, +7 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 7 tools
Each tool maps to a distinct step in the translation workflow: quoting, ordering, single-task lookup, task listing, downloading, account balance, and engine discovery. The overlap between fanbei_quote and fanbei_translate_document is intentional (estimate vs. purchase) and clearly described.
All tools use a consistent fanbei_ snake_case prefix, but the patterns mix verb_noun (translate_document, list_tasks, task_status) with bare nouns (quote, download, account, engines). This is readable and predictable overall, with only minor deviations.
Seven tools is well-scoped for a document translation service and each tool covers a necessary workflow step. There is no obvious redundancy or filler.
The core lifecycle is covered: quote, order, check status, list tasks, and download, plus account and engine metadata. However, there is no cancel, delete, or restore tool despite fanbei_list_tasks exposing a trash scope, and no retry for failed tasks, which are minor but real dead ends.
Maintenance
Related MCP Connectors
Convert, create and edit Word, PowerPoint, Excel and PDF documents from any AI agent.
Hosted document tools for agents: extract, parse, classify, convert, OCR, sign, and redact.
Document conversion and OCR for AI agents: PDF, Office docs, images to text.
Agentic localization for documents and media, with quality checks and optional human review.
Related MCP Servers
- AlicenseAqualityDmaintenanceEnables parsing and extraction of content from various document formats (PDF, Word, Excel, PowerPoint) into Markdown format using the Niutrans document API.15MIT
- AlicenseNot gradedqualityBmaintenanceEnables document merging, splitting, and info retrieval for PDF, DOCX, XLSX, PPTX, and 30+ formats via AI agents.MIT
- AlicenseNot gradedqualityCmaintenanceEnables agents to generate PowerPoint, Excel, Word, and Markdown files, review DOCX files with comments, reuse catalog images, and receive structured download metadata.MIT
- AlicenseNot gradedqualityBmaintenanceEnables MCP agents to inspect and edit Word, Excel, PowerPoint, PDF, and text documents locally while preserving formatting, including in-place text replacement, HTML round-tripping of whole files or individual worksheets, slides, or PDF page ranges, and filling Word form fields.MIT