As AI agents proliferate across platforms and vendors, organizations increasingly lack visibility into which agents exist, what they do, and whether they operate safely. SAP AI Agent Hub is a vendor-agnostic AI governance solution that addresses this directly: it gives enterprise architects, IT administrators, and AI Centers of Excellence a single, structured system of record for all AI assets in the organization, regardless of where they were built or which platform they run on. It covers AI agents, large language models (LLMs), and MCP servers across your entire IT landscape.
SAP AI Agent Hub is available for Business AI Platform customers who get Joule Studio and/ or AI Units, and as part of SAP LeanIX, SAP’s enterprise architecture management solution.
Discover and inventory your AI assets
SAP AI Agent Hub automatically discovers AI agents, LLMs, and MCP servers from the platforms where they are built and deployed. Supported sources include Microsoft Entra ID, Google Agent Engine, Amazon Bedrock, Microsoft Azure API Center, and, new with this release, ServiceNow and SAP AI Core. All discovered assets are organized in a centralized inventory with classification, status, and architecture context.
Evaluate and verify AI assets
Once discovered, AI agents and MCP servers enter your inventory in an unverified state. Structured assessments capture key governance attributes for each asset, including AI risk rating, build type, and compliance mappings. Architecture decisions can be attached directly to each fact sheet, giving you a place to document the rationale behind agent adoption, platform choices, and compliance boundaries — and informing whether an asset should be verified or not. Based on this assessment, you can mark any asset as verified, rejected, or unverified, with a verification badge appearing on each fact sheet. Going forward, the badge will be picked up by the Joule Studio runtime to inform which agents are approved to run in production, and through an integration with SAP Integration Suite, it will indicate which MCP servers are verified for use by your developers.
Outlook
We are planning to further enhance SAP AI Agent Hub for it to become the central command center of AI assets, surfacing relevant information from diverse sources and ensuring a seamless user journey to access relevant data:
Observability will give you visibility into the runtime behavior, performance, and usage of your AI agents and models in production, with drill-down into individual sessions via SAP Cloud ALM.
Verification status of agents and MCP servers will be visible in Joule Studio and SAP Integration Suite and indicate safe usage.
Identity and access control, built on SAP Cloud Identity Services, will enable you to centrally manage access, identities, and policies for AI agents across your organization.
Agent mining, via an integration with SAP Signavio, will help you understand how agents execute tasks across defined process steps, measuring conformance and coverage.
Org chart and skills mapping via SAP SuccessFactors will allow you to map AI agent deployment to business units and organizational structures, helping you understand workforce impact as agent adoption scales.
With this, we continue to bring more visibility to your AI ecosystem to give you the confidence to scale your AI strategies.
As AI agents proliferate across platforms and vendors, organizations increasingly lack visibility into which agents exist, what they do, and whether they operate safely. SAP AI Agent Hub is a vendor-agnostic AI governance solution that addresses this directly: it gives enterprise architects, IT administrators, and AI Centers of Excellence a single, structured system of record for all AI assets in the organization, regardless of where they were built or which platform they run on. It covers AI agents, large language models (LLMs), and MCP servers across your entire IT landscape.
SAP AI Agent Hub is available for Business AI Platform customers who get Joule Studio and/ or AI Units, and as part of SAP LeanIX, SAP’s enterprise architecture management solution.
Discover and inventory your AI assets
SAP AI Agent Hub automatically discovers AI agents, LLMs, and MCP servers from the platforms where they are built and deployed. Supported sources include Microsoft Entra ID, Google Agent Engine, Amazon Bedrock, Microsoft Azure API Center, and, new with this release, ServiceNow and SAP AI Core. All discovered assets are organized in a centralized inventory with classification, status, and architecture context.
Evaluate and verify AI assets
Once discovered, AI agents and MCP servers enter your inventory in an unverified state. Structured assessments capture key governance attributes for each asset, including AI risk rating, build type, and compliance mappings. Architecture decisions can be attached directly to each fact sheet, giving you a place to document the rationale behind agent adoption, platform choices, and compliance boundaries — and informing whether an asset should be verified or not. Based on this assessment, you can mark any asset as verified, rejected, or unverified, with a verification badge appearing on each fact sheet. Going forward, the badge will be picked up by the Joule Studio runtime to inform which agents are approved to run in production, and through an integration with SAP Integration Suite, it will indicate which MCP servers are verified for use by your developers.
Outlook
We are planning to further enhance SAP AI Agent Hub for it to become the central command center of AI assets, surfacing relevant information from diverse sources and ensuring a seamless user journey to access relevant data:
Observability will give you visibility into the runtime behavior, performance, and usage of your AI agents and models in production, with drill-down into individual sessions via SAP Cloud ALM.
Verification status of agents and MCP servers will be visible in Joule Studio and SAP Integration Suite and indicate safe usage.
Identity and access control, built on SAP Cloud Identity Services, will enable you to centrally manage access, identities, and policies for AI agents across your organization.
Agent mining, via an integration with SAP Signavio, will help you understand how agents execute tasks across defined process steps, measuring conformance and coverage.
Org chart and skills mapping via SAP SuccessFactors will allow you to map AI agent deployment to business units and organizational structures, helping you understand workforce impact as agent adoption scales.
With this, we continue to bring more visibility to your AI ecosystem to give you the confidence to scale your AI strategies. Read More Technology Blog Posts by SAP articles
#SAP
#SAPTechnologyblog
