一个模型上下文协议(MCP)服务器,用于将SAP系统与克劳德等AI助手使用OData REST API进行集成. 这个服务器提供了连接SAP OData服务,查询实体集,执行CRUD操作,并调用OData功能的工具.

□ 特性

• SAP OData连接:通过OData REST API与SAP系统连接
• 智能连接处理: 妥善处理SAP OData URL结构和404个答复
• 服务发现:通过目录或共同事务测试自动发现现有的OData服务
• 实体设置查询: 查询任何带有过滤、排序和页码的 OData 实体
• ** CRUD 操作**:在 OData 实体上创建、读取、更新并删除操作
• ** 功能导入**:执行OData函数导入和自定义函数
• CSRF Token处理:用于安全操作的自动CSRF标志管理
• ** 模块架构**:清洁、可维护的类型脚本编码库,同时分离关切

□ 先决条件

• ** 第18号节点。
• ** 启用具有OData服务的SAP系统**
• ** SAP OData端点的网络接入**
• ** 附有适当授权的SAP用户证书**

SAP OData MCP Server

A Model Context Protocol (MCP) server for integrating SAP systems with AI assistants like Claude using OData REST APIs. This server provides tools for connecting to SAP OData services, querying entity sets, executing CRUD operations, and calling OData functions.

Features

• SAP OData Connectivity: Connect to SAP systems via OData REST APIs
• Smart Connection Handling: Properly handles SAP OData URL structures and 404 responses
• Service Discovery: Automatically discover available OData services via catalog or common service testing
• Entity Set Queries: Query any OData entity set with filtering, sorting, and pagination
• CRUD Operations: Create, Read, Update, and Delete operations on OData entities
• Function Imports: Execute OData function imports and custom functions
• CSRF Token Handling: Automatic CSRF token management for secure operations
• Modular Architecture: Clean, maintainable TypeScript codebase with separation of concerns

Prerequisites

• Node.js 18+
• SAP system with OData services enabled
• Network access to SAP OData endpoints
• SAP user credentials with appropriate authorizations

⚠️ Advantage: No SAP RFC SDK installation required! Uses standard HTTP/REST APIs.

Installation

Quick Setup

1. Create the project:

mkdir sap-odata-mcp-server
cd sap-odata-mcp-server
mkdir src

2. Copy the source files from the artifacts to your src/ directory:

   • src/index.ts - Entry point
   • src/server.ts - MCP server setup
   • src/handlers.ts - Request handlers
   • src/odata-client.ts - SAP OData client
   • src/tool-definitions.ts - Tool definitions
   • src/types.ts - TypeScript types

3. Copy configuration files:

   • package.json - Dependencies and scripts
   • tsconfig.json - TypeScript configuration
   • .env.example - Environment variables template

4. Install dependencies:

npm install

5. Configure environment:

cp .env.example .env
# Edit .env with your SAP details

6. Build the project:

npm run build

Configuration

Environment Variables

Create a .env file with your SAP system details:

# Required SAP OData Configuration
SAP_ODATA_BASE_URL=https://your-sap-host:8000/sap/opu/odata/sap/
SAP_USERNAME=your-sap-username
SAP_PASSWORD=your-sap-password

# Optional Configuration
SAP_CLIENT=100
SAP_TIMEOUT=30000
SAP_VALIDATE_SSL=false  # for development with self-signed certificates
SAP_ENABLE_CSRF=true

Claude Desktop Integration

Add to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "sap-odata": {
      "command": "node",
      "args": ["/full/path/to/your/sap-odata-mcp-server/dist/index.js"],
      "env": {