云启智联 MCP Server
OfficialThis server provides AI-powered document parsing capabilities for financial and general documents, allowing AI clients to extract structured data from images or PDFs.
parse_bank_receipt: Parse bank receipt images or PDFs, with automatic cropping of multiple receipts per page. Extracts structured fields such as transaction date, summary, amount, account numbers, account holder names, bank names, and transaction reference numbers.parse_bank_statement: Parse bank statement images or PDFs, returning structured account information and transaction details.parse_invoice: Parse VAT invoice images or PDFs, extracting invoice code, number, issue date, amounts, and buyer/seller information.parse_document: General document parsing to extract structured text content from various PDFs or images.get_task_result: Query the results of previously submitted asynchronous parsing tasks using a task ID.
All tools require a 云启智联 API Key and accept either a local file path or a remote file URL as input. An optional callback URL can be provided for asynchronous task notifications.
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., "@云启智联 MCP Server 解析这张银行回单 /tmp/receipt.pdf"
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.
云启智联 MCP Server
云启智联AI文档解析服务的 Model Context Protocol (MCP) 服务器,让 Claude、Cursor、Windsurf 等支持 MCP 的 AI 客户端直接调用票据识别能力。
功能
parse_bank_receipt— 银行回单解析(支持每页多张自动裁剪)parse_bank_statement— 银行对账单解析parse_invoice— 增值税发票解析parse_document— 通用文档解析get_task_result— 查询异步任务结果
Related MCP server: @accos/mcp-server
前置要求
Python 3.10+
云启智联 API Key(从 http://8.135.62.13:5000 注册获取)
安装
pip install yqzl-mcp-server或从源码安装:
git clone https://github.com/yunqi-zhilian/yqzl-mcp-server.git
cd yqzl-mcp-server
pip install -e .启动
安装后,如果 yqzl-mcp-server 命令可用,直接运行:
yqzl-mcp-server部分环境(尤其是 Windows)安装后脚本目录可能不在 PATH 中,导致命令找不到。此时可使用 Python 模块方式启动,效果完全相同:
python -m yqzl_mcp_server.server在 Claude Desktop 中配置
编辑 ~/Library/Application Support/Claude/claude_desktop_config.json(macOS)或 %APPDATA%\Claude\claude_desktop_config.json(Windows):
{
"mcpServers": {
"yqzl-ai": {
"command": "yqzl-mcp-server",
"env": {
"YQZL_API_BASE_URL": "http://8.135.62.13:5000/AIService"
}
}
}
}如果系统提示找不到 yqzl-mcp-server 命令(常见于 Windows),可将 command 改为使用 Python 模块方式启动:
{
"mcpServers": {
"yqzl-ai": {
"command": "python",
"args": ["-m", "yqzl_mcp_server.server"],
"env": {
"YQZL_API_BASE_URL": "http://8.135.62.13:5000/AIService"
}
}
}
}配置完成后重启 Claude Desktop,即可在对话中让 Claude 调用票据识别工具。
在 Cursor 中配置
打开 Cursor Settings → MCP,添加 server:
Type:
stdioName:
yqzl-aiCommand:
yqzl-mcp-server
如果命令找不到,可改用:
Command:
pythonArgs:
-m yqzl_mcp_server.server
使用示例
对 AI 说:
请用云启智联解析这张银行回单
/Users/xxx/receipt.pdf,API Key 是as-xxxxxxxx。
AI 会自动调用 parse_bank_receipt 工具并返回结构化结果。
计费说明
MCP Server 不掌握计费逻辑。所有调用都通过你的云启智联 API Key 进行,按量从你在云启智联平台的余额中扣费。具体价格见官网。
自定义上游地址
默认使用 http://8.135.62.13:5000/AIService。域名备案完成后,可通过环境变量切换:
export YQZL_API_BASE_URL="https://yunqi-zhilian.com/AIService"开发
pip install -e ".[dev]"
python -m yqzl_mcp_server.server许可证
MIT
Available Tools
5 toolsget_task_resultB
根据 task_id 查询异步解析任务的结果。
Inputs: api_key (required), task_id (required).
parse_bank_receiptA
解析银行回单图片或 PDF,支持每页多张回单自动裁剪为单张。返回结构化字段:交易日期、摘要、金额、收支账号户名、银行、交易流水号等。
Inputs: api_key (required), file_path, file_url, callback_url.
parse_bank_statementB
解析银行对账单图片或 PDF,返回账户信息、交易明细等结构化数据。
Inputs: api_key (required), file_path, file_url, callback_url.
parse_documentC
通用文档解析,提取 PDF/图片中的结构化文本内容。
Inputs: api_key (required), file_path, file_url, callback_url.
parse_invoiceB
解析增值税发票图片或 PDF,返回发票代码、发票号码、开票日期、金额、购销方信息等。
Inputs: api_key (required), file_path, file_url, callback_url.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 5 tools
Each parse tool targets a distinct document type (bank receipt, bank statement, invoice, generic document), and get_task_result is clearly separate for async task management. No overlap.
Most tools follow parse_* pattern, but get_task_result uses get_ instead. However, the action difference justifies the variation, and naming is otherwise consistent snake_case.
5 tools is appropriate for a document parsing server, covering major document types and adding a task result retrieval tool without bloat.
Core document parsing types (receipt, statement, invoice, generic) are covered, and async result retrieval is included. Missing potentially other document types like contracts, but not a significant gap for the stated purpose.
Maintenance
Related MCP Connectors
Generate professional receipt PDFs and structured receipt data with AI-ready MCP tools.
Remote MCP + OAuthで、MCP対応AIから見積書・請求書等を作成・編集・変換。PDF生成、共有URL、品目マスタ、学習結果の保存にも対応。
A paid remote MCP for AI SDK data query MCP, built to return verdicts, receipts, usage logs, and aud
A paid remote MCP for AI SDK MCP gateway registry, built to return verdicts, receipts, usage logs, a
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceMCP server for ReceiptConverter that allows AI assistants to parse any receipt or invoice image/PDF into structured JSON with a single tool call.12 npmMIT
- AlicenseNot gradedqualityDmaintenanceConnects AI agents to ACCOS for OCR on Thai receipts, tax invoices, and withholding tax certificates via MCP.Apache 2.0
- FlicenseNot gradedqualityDmaintenanceMCP server for document conversion and processing, enabling AI assistants to handle various document formats.-
- FlicenseNot gradedqualityCmaintenanceEnables AI assistants to convert and process various document formats through an MCP server interface.-