What's New

AWS has announced the General Availability of the AWS for SAP MCP Server on Amazon Bedrock AgentCore, a purpose-built integration layer that connects AI agents directly to SAP ERP systems at enterprise scale. Built on the Model Context Protocol (MCP) and SAP's Open Data Protocol (OData), the solution enables AI agents to perform full CRUD operations on SAP business objects — including sales orders, purchase orders, materials, and finance documents — across SAP S/4HANA and SAP ECC environments. The server ships as a container image at no cost and can be deployed in minutes via CloudFormation templates.

How It Works

• The AWS for SAP MCP Server runs on the fully managed Amazon Bedrock AgentCore Runtime, which provides the underlying compute and lifecycle management.
• It exposes SAP ERP capabilities to AI agents through the standardized MCP interface, translating agent requests into SAP OData API calls that interact with SAP business processes spanning finance, procurement, logistics, and supply chain.
• Security is handled via a dual-layer authentication model through AgentCore Identity, supporting OAuth 2.0, while session isolation and private connectivity ensure data segregation between agent sessions.
• Dynamic service catalog discovery allows agents to introspect available SAP services at runtime without hardcoded configurations.
• Observability is provided natively through Amazon CloudWatch, giving operators full telemetry into agent actions and SAP interactions.
• Compatible MCP clients include Amazon Q (referred to as "Amazon Quick" in the announcement), Strands SDK-based custom agents, and SAP's own Joule assistant.

Why It's Important

• SAP ERP systems are the operational backbone for thousands of enterprises, yet integrating them with modern AI agents has historically required complex, bespoke middleware development with significant security and compliance overhead.
• This announcement removes that barrier by providing a production-ready, managed integration path that maintains enterprise-grade security and auditability.
• Organizations can now automate high-value workflows — such as Procure-to-Pay, order management, and financial reconciliation — using AI agents without building custom connectors or managing integration infrastructure.
• The CloudWatch-based observability layer is particularly significant for enterprise governance, as it provides auditable records of every agent action taken against SAP data, which is critical for compliance in regulated industries.

How It's Different

• Prior to this release, connecting AI agents to SAP systems typically required custom OData client development, SAP BTP (Business Technology Platform) middleware, or third-party iPaaS solutions, all of which demanded significant engineering effort and introduced additional operational complexity.
• Existing SAP integration patterns on AWS (e.g., using AWS Lambda with SAP OData APIs directly) lacked standardized agent-facing interfaces, session management, and built-in identity federation.
• The MCP standard differentiates this solution by providing a protocol-level abstraction that any MCP-compatible agent or client can consume without SAP-specific knowledge.
• The managed runtime on AgentCore further distinguishes it from self-hosted alternatives by eliminating infrastructure provisioning, patching, and scaling concerns.
• The zero-cost container image model also contrasts with commercial middleware licensing fees common in the SAP integration ecosystem.

When to Prefer It

• This solution is the clear choice for organizations already running SAP S/4HANA or SAP ECC on AWS that want to build or extend AI agent workflows against live SAP business data without investing in custom integration development.
• It is particularly well-suited for use cases involving multi-step agentic workflows — such as automated procurement approvals, AI-driven financial close assistance, or supply chain exception handling — where agents need to both read context and write back to SAP transactional systems.
• Teams using Amazon Q Business, Strands SDK, or SAP Joule as their agent runtime should prefer this over custom integrations due to native MCP client compatibility.
• Organizations with strict security and compliance requirements will benefit from the built-in OAuth 2.0 support, session isolation, and CloudWatch audit trails.
• If your SAP landscape is on-premises or hosted outside AWS, additional network connectivity considerations (e.g., AWS Direct Connect or VPN) will apply but the solution remains viable.

Availability

• The AWS for SAP MCP Server is now Generally Available as of May 1, 2026.
• It supports both SAP S/4HANA and SAP ECC as target ERP systems.
• The server is delivered as a container image at no additional cost beyond underlying AgentCore Runtime and CloudWatch usage charges.
• Deployment is facilitated through AWS CloudFormation templates, requiring no custom infrastructure management.
• Specific supported AWS regions are not enumerated in the announcement, so users should consult the official AWS for SAP MCP Server documentation page for the current regional availability matrix.
• Early adopter deployments by Fortescue, Harman International, and PLDT confirm production readiness across diverse enterprise use cases.