SAP-Claude Integration via MCP (Model Context Protocol)
@CostingGeek
SAP-Claude Integration via MCP (Model Context Protocol) について
An MCP server to retrieve information like Sales Orders from the SAP Graph API Sandbox
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"sap-mcp": {
"command": "uv",
"args": [
"init",
"saptools"
]
}
}
}ツール
ツールは検出されませんでした
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
概要
What is SAP-Claude Integration via MCP (Model Context Protocol)?
SAP-Claude Integration via MCP is a Python-based solution that enables bidirectional communication between Claude (Anthropic’s large language model) and SAP systems through the SAP Graph API. It allows Claude to retrieve and process real-time SAP business data using the Model Context Protocol.
How to use SAP-Claude Integration via MCP (Model Context Protocol)?
Install Python 3.8+, set up uv and a Python virtual environment, then clone the provided saptools.py file. Configure Claude Desktop by editing claude_desktop_config.json with the correct path to the server folder and restart Claude. Once connected, you can ask Claude prompts like “Give me the last 10 sales orders” or “What is the average amount of the last 5 sales orders?”
Key features of SAP-Claude Integration via MCP (Model Context Protocol)
- Connect Claude to SAP systems with minimal setup.
- Retrieve real-time sales order data via SAP Graph API.
- Extend the integration with additional SAP entities (Customers, Products, etc.).
- Process SAP data with Claude’s full LLM capabilities.
- Uses FastMCP to expose SAP functionality as tools.
- Currently provides the
get_sales_orderstool.
Use cases of SAP-Claude Integration via MCP (Model Context Protocol)
- Query real-time sales orders using natural language.
- Analyze sales data, such as calculating average order amounts.
- Extend to retrieve customer, product, or other SAP entity data.
- Integrate AI-driven insights into SAP business workflows.
FAQ from SAP-Claude Integration via MCP (Model Context Protocol)
What are the prerequisites to run this integration?
Python 3.8+, a SAP API key (sandbox or production), Claude Desktop, and a basic understanding of MCP and the SAP Graph API.
How do I obtain a SAP API key?
Log into the SAP Graph Navigator, go to your profile settings under Settings > API Settings, and save the API key as a system environment variable named SAP_API_KEY.
What tools are currently available?
The integration currently provides one tool: get_sales_orders. It retrieves sales orders from the SAP Graph API sandbox, with an optional top parameter (default 3) and an optional api_key parameter.
How do I connect Claude Desktop to the MCP server?
Edit the claude_desktop_config.json file (usually in %APPDATA%\Claude) to add the server configuration with uv run command pointing to your saptools.py file. Then completely close and restart Claude Desktop.
Is this integration production-ready?
No, the current version is a proof-of-concept limited to the get_sales_orders tool, with plans to support more complex interactions like querying orders for a specific customer.
「AI とエージェント」の他のコンテンツ
Gemini MCP Server
aliargunMCP server implementation for Google's Gemini API
Mcp Agent
lastmile-aiBuild effective agents using Model Context Protocol and simple workflow patterns
mcp-hfspace MCP Server 🤗
evalstateMCP Server to Use HuggingFace spaces, easy configuration and Claude Desktop mode.
欢迎来到 智言平台
Shy2593666979AgentChat 是一个基于 LLM 的智能体交流平台,内置默认 Agent 并支持用户自定义 Agent。通过多轮对话和任务协作,Agent 可以理解并协助完成复杂任务。项目集成 LangChain、Function Call、MCP 协议、RAG、Memory、HITL、Skill、Milvus 和 ElasticSearch 等技术,实现高效的知识检索与工具调用,使用 FastAPI 构建高性能后端服务。
MCP-LLM Bridge
patruffBridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools
コメント