Model Context Protocol (MCP) Server Integration
Model Context Protocol (MCP) is an open standard that defines how AI models and agents can discover, understand, and interact with tools, data sources, and services. Developers use it to expose capabilities — such as SAP BTP account administration or the hana-ml Python library — as MCP endpoints that AI agents can consume in a standardized way. For example, the HANA AI Toolkit local MCP server turns the hana-ml library into an MCP-accessible endpoint, enabling third-party agents and AI models to invoke its functionality without custom integrations.
Tutorials that teach this
- Tutorial Use Joule Studio to Extend the AP Invoice Agent using Agent Extensions
- Tutorial Joule Studio CodeJam - Introduction
- Tutorial Add MCP capabilities to a CAP service
- Tutorial Set Up MCP Server for MDK Development
- Tutorial Create Your First MDK Application Using MCP Server
- Tutorial 8 - Add Subagent to Convert Currency Using MCP Server
- Tutorial 🟡 Devtoberfest 2025 - Week 4 - Developer Advocates Discuss
Docs explaining this concept
- Doc A2A and MCP for Interoperability
- Doc Agentic AI & AI Agents
- Doc Agentic AI Foundation
- Doc Global Standards for Agentic AI
- Doc HANA AI Toolkit - Local MCP Server
- Doc Integration, Security, Ethics & Governance
- Doc Microsoft Copilot Studio and the MCP Gateway in SAP Integration Suite
- Doc Third-Party MCP Access to SAP Solutions
Videos teaching this
Code samples embodying this
Prerequisites
- Concept IDE and AI Integration for Developers The provided sources do not contain sufficient information to write a grounded definition for an undefined concept. No factual claims about a specific SAP developer concept can be made based solely on a Devtoberfest 2025 tutorial title with no accompanying content.
- Concept npm Package Management The concept provided is `undefined`, and the supplied sources do not contain sufficient information to produce a grounded, accurate definition for an unnamed or undefined SAP developer concept. Please provide a valid concept name so a precise and source-grounded definition can be written.
- Concept CAP OData V4 Service Development The concept provided is **undefined**, and the supplied sources do not describe a specific, identifiable SAP developer concept that can be accurately defined. No definition can be responsibly written without a clearly stated concept name and supporting source content.
- Concept JSON-RPC Protocol The **Model Context Protocol (MCP)** is a protocol that enables AI models to interact with external services and tools in a structured way. Developers use it to expose capabilities — such as those from a CAP service — so that AI assistants can discover and invoke them at runtime. By [adding MCP capabilities to a CAP service](https://developers.sap.com) and [examining the protocol details on startup](https://developers.sap.com), developers can integrate their backend services directly into AI-driven workflows.
- Concept Large Language Models (LLMs) The concept provided is **undefined**, and the supplied sources do not describe a specific, identifiable SAP developer concept by that name. Based on the available snippets — which cover topics such as [Retrieval Augmented Generation (RAG)](https://architecture.learning.sap.com/docs/ref-arch/RA0005/3-retrieval-augmented-generation/readme), prompting LLMs in the generative AI hub, and using custom or small language models on SAP AI Core — no definition can be responsibly constructed for an unnamed or undefined concept. Please provide a valid concept name so that an accurate, source-grounded definition can be written.
- Concept SAP Mobile Services Account Setup and Configuration No concept was provided, and the supplied sources do not converge on a single, identifiable SAP developer concept that can be defined. A meaningful definition cannot be written without a specified concept to define.
- Concept CAP Framework (Cloud Application Programming Model) The concept name was not provided, so a precise definition cannot be determined. Based on the available sources, [SAP Business Technology Platform (SAP BTP)](https://help.sap.com/docs/btp/sap-business-technology-platform/c2fec62b49fa43b8bd945c85ecc2e5bd?locale=en-US&state=PRODUCTION&version=Cloud) is a platform on which developers build and run business applications, choosing from development environments, tools, APIs, and [programming models](https://help.sap.com/docs/btp/sap-business-technology-platform/042061da2e964b7eab3c7ec2a5cc7ece?locale=en-US&state=PRODUCTION&version=Cloud) such as ABAP Cloud and the SAP Cloud Application Programming Model (CAP). Developers use it to create full-stack business applications, following best practices to select the approach that best suits their requirements.
- Concept CAP AI Plugin and MCP Integration
- Concept Node.js Runtime Installation The concept name provided is "undefined," which does not correspond to a identifiable SAP developer concept supported by the given source snippets. The sources cover topics such as connecting to SAP HANA and Data Lake Relational Engine via Node.js drivers and setting up local development environments for UI5 and Node.js, but none of them define or describe a concept called "undefined." A valid concept name is required to produce an accurate, source-grounded definition.
- Concept CAP Service Modeling The concept name provided is "undefined," so no specific SAP developer concept can be identified or defined from the available sources. The sources cover topics such as [CAP services with Node.js](https://developers.sap.com/tutorials/appstudio-cap-nodejs-create.html) and SAP HANA stored procedures, but without a valid concept name, a grounded definition cannot be produced. Please provide a specific concept name to generate an accurate reference definition.
- Concept Agentic AI Architectures Agentic AI architectures are design patterns in which AI systems autonomously plan and execute multi-step tasks, often orchestrating tools, APIs, and other services to achieve a goal. Developers working within the SAP ecosystem can use [SAP's AI Golden Path](https://architecture.learning.sap.com/docs/golden-path/ai-golden-path/readme) as a starting point for building and structuring such AI applications, following SAP's recommended practices and architectural guidance.
Concepts that build on this
- Concept LLM Tool Use / Function Calling in Agents
- Concept Devtoberfest Data & AI Sessions Overview
- Concept Developer Tools in the 21st Century
- Concept SAP AI Skills Library
- Concept MDK Project Setup in SAP Build
- Concept Agentic AI Coding
- Concept Agent Extension Development with Post-Hooks and Instructions
- Concept Joule Studio Project Release and Deployment
- Concept Devtoberfest 2025 Week 2 CAP Sessions
- Concept MCP Inspector TUI Client
- Concept MCP-Grounded AI Agentic Development
- Concept Model Context Protocol (MCP) for Agentic SAP Integration