NewSAP Business Technology Platform, also named SAP Business AI Platform
In the last 7 days
57 changes
Today
Joule Studioがリリースされたので早速CAPアプリを自然言語で操作するカスタムスキルを構築してみたpublishedA walkthrough of building a custom Joule StudioJoule StudioSAP’s managed development environment for building custom agents, applications and workflows from business intent, in six phases from intent and specification to deployment on an SAP-managed runtime.sap official, umbrella skill on a CAP bookshop app:
cds add sample and cds add mta scaffold the project and its deployment descriptor, cds compile srv/ --to xsuaa generates the role template behind the @requires: 'admin' check on AdminService from the SAP Authorization and Trust Management serviceSAP Authorization and Trust Management serviceThe OAuth server behind a BTP application: issues the JWT the app checks, holds the scopes and role collections it checks against, and brokers trust to an identity provider. Bound, never installed.sap official, btp (XSUAA), and mbt build packages the app for deployment to the SAP BTP, Cloud Foundry RuntimeSAP BTP, Cloud Foundry RuntimeThe Cloud Foundry runtime of SAP BTP: memory for applications in a Cloud Foundry org, run from SAP's Java, Node.js and Python buildpacks or community buildpacks and managed with the standard Cloud Foundry tools.sap official, btp. The skill reaches the deployed ODataODataExposing data and operations through OData, including the service model, metadata and protocol versions clients require. Version differences affect which clients can consume a service and which query features they can use.SAP and community, narrow by design API through the SAP Destination serviceSAP Destination serviceNamed, credentialed addresses of other systems, resolved at runtime — so an application holds no URL and no secret of its own, and the same code reaches a different system in every landscape.sap official, btp, configured with the OAuth2 Client Credentials grant using the client ID, secret and token endpoint from the app's XSUAA service key, so Joule answers a natural-language question about a book by its ID.Agent toolingAgent access to your codecoverage by ecosystem, coverage by flavor revisedReassessed against the graph (check:capabilities): removed the tool count and the two absence claims about who has built something.Agent access to your code
Agent evalscore coverage, coverage by ecosystem revisedReassessed against the graph (check:capabilities): albertmink-agentic-abap-evals, an independent set of ABAP tasks and a grader for coding agents, was entered after the block was written; the count is dropped and the second kind named. State unchanged.Agent evals
Agent guidancecoverage by ecosystem, coverage by flavor revisedReassessed against the graph (check:capabilities): docs/WRITING.md §3 rules out provider counts and absence claims in every capability field.Agent guidance
Show 12 more
Agent toolingcore coverage, coverage by flavor, coverage by ecosystem revisedReassessed against the graph (check:capabilities): sap-joule-studio and sap-joule, both sap-official, now provide this capability behind the BTP door, so the note's claim about every entry no longer held. The state stays core-only: no provider of another origin counts for the door.Agent tooling
Agent workspacescore coverage revisedReassessed against the graph (check:capabilities): Joule Work is a sap-official provider serving BTP, so the core is present; the note's absence claim is dropped.Agent workspaces
Attachmentscoverage by ecosystem, coverage by flavor revisedReassessed against the graph (check:capabilities): drops a count-like contrast with a plugin the flavor does not have.Attachments
Cataloguing and lineagecore coverage revisedReassessed against the graph (check:capabilities): a newer community server publishes Delta shares as data products in Business Data Cloud, a kind the text did not name.Cataloguing and lineage
Database connectorscoverage by ecosystem revisedReassessed against the graph (check:capabilities): the note counted one entry behind the door; it now names the driver and SAP HANA, which now provides the capability on this door. State unchanged.Database connectors
Document extractioncoverage by ecosystem revisedReassessed against the graph (check:capabilities): removed the provider-count and absence claims (docs/WRITING.md §3); the states are unchanged.Document extraction
LLM integrationcoverage by ecosystem, core coverage, coverage by flavor revisedReassessed against the graph (check:capabilities): the note claimed who provides the capability behind this flavor, which moves with the corpus. The flavor state is unchanged.LLM integration
Loading a warehousecore coverage, coverage by ecosystem revisedReassessed against the graph (check:capabilities): ERPL is a provider the text never named, and it loads from ABAP systems outside BW routines. State unchanged.Loading a warehouse
Predictive modelscore coverage revisedReassessed against the graph (check:capabilities): the note explained the state by the batch it was written from and counted repositories; it now names the kinds of SAP service that have since arrived. The state stays core-only: both are sap-official services on the btp door.Predictive models
Prompt managementcoverage by ecosystem, core coverage revisedReassessed against the graph (check:capabilities): the note was a claim about who provides, which the option list derives.Prompt management
Retrieval-augmented generationcoverage by ecosystem, core coverage revisedReassessed against the graph (check:capabilities): a CAP-door provider reaches Document Grounding through the LangChain vector store interface, a shape the two plugins do not cover.Retrieval-augmented generation
Tool governancedescription, core coverage, coverage by ecosystem, framing by ecosystem, coverage by flavor revisedReassessed against the graph (check:capabilities): a count of providers removed.Tool governance
Yesterday
SAP CAP With Prostgres Serice in BTPpublishedA step-by-step walkthrough of a CAP bookshop project on @cap-js/postgres@cap-js/postgresThe PostgreSQL database service — the production option for running CAP off HANA, sharing the base service with the other three.sap official, cap, provisioning PostgreSQL on SAP BTP, hyperscaler optionPostgreSQL on SAP BTP, hyperscaler optionA managed PostgreSQL from the underlying hyperscaler — AWS, Azure, Google Cloud or Alicloud — consumed through a BTP service binding.sap official, btp's trial plan through
mta.yaml, building the MTA archive and deploying it to Cloud Foundry by hand, then inspecting the deployed database with pgAdmin over an SSH tunnel. It updates an earlier post by Mike ZaschkaMike Zaschkamaintains 6 entries, created 12 pieces, contributed to 9 entriesindependent on the same topic to the package CAP now ships by default, though the manual build-and-deploy sequence it shows predates 
cds up folding those steps into one command by a day.Database connectorsAgent access to your codecoverage by ecosystem revisedRewritten at the maintainer's request (2026-10-25): what SAP ships named whole, and what the community adds written from what its ABAP MCP servers state about how they run, connect and sign in.Agent access to your code
Sat 24
btp-agentadded to the atlasLangGraph agent and UI5 chat screen over four BTP MCP servers
SAP-samples/hana-cli-claude-pluginadded to the atlasClaude Code plugin registering hana-cli's MCP serverAgent access to your systems
Using BTP Document Extraction Service into actual App developmentpublishedWalks through calling SAP Document AI's REST API directly: getting a client-credentials bearer token from the service key's UAA, then POSTing to /document/jobs with a file and an options body naming the document type and the header fields and line items to extract (paymentTerms, documentDate, taxId, description for an invoice), and polling the returned job id until the status resolves. Covers the trial-account booster and its own UI for uploading a document and testing extraction before moving to Postman and a small Python script that lists jobs over the same bearer token.Document extraction
SAP HANA Cloudprovides revisedThe capability was added on 2026-10-24 for administering a database instance; SAP HANA Cloud ships its administration tools, so it is the core answer on BTP.Database connectors
Show 1 more
SAP-samples/hana-developer-cli-tool-exampleprovides, serves ecosystems, domain, provides revisedIts performance analysis, privilege grants and backup and recovery commands had no capability to sit under; Database administration was added for them the same day.Database administration
Fri 23
Cloud Integration with Document AIpublishedWalks through connecting Cloud IntegrationCloud IntegrationBuilds and runs integration flows that process, route and map messages between cloud and on-premise systems, SAP and non-SAP, through prebuilt adapters; available only under an SAP Integration Suite plan.sap official, btp to SAP Document AISAP Document AIExtracts header fields and line items from invoices, payment advices, purchase orders and custom document types, as structured data over a REST API — the extraction is the service's, the schema can be yours.sap official, btp: an iFlow posts a prescription PDF to Document AI's job API, polls the job endpoint until it completes, and a groovy script reads back the result, keeping only the fields above a stated 60% confidence threshold. The scenario is a hospital extracting a new patient's past illnesses, allergies and blood group from uploaded prescriptions, matched against a custom extraction schema.Document extraction
BTP MCP Serverserves ecosystems revisedThe tag came in with the entry on 2026-09-10 and nothing sourced it: the README names BTP Core Services and the proxy it runs on, never CAP. It read as a routing leftover from when the MCP servers capability lived on the CAP door; the entry stays on BTP, its home.Agent access to your systems
hana-mcp-serverserves ecosystems revisedThe tag came in with the entry on 2026-09-10 and nothing sourced it: the README names HANA and no CAP stack. Removed with the other unsourced tags from that sweep; the entry stays on BTP, its home.Agent access to your systems
odata-mcp-proxyserves ecosystems revisedThe tag came in with the entry on 2026-09-10 and nothing sourced it: the README describes a config-driven proxy on BTP Cloud Foundry and names no CAP stack. Removed with its sibling configs; the entry stays on BTP, its home.Agent access to your systems
Show 3 more
SAP Analytics Cloudlinks revisedSAP moved the What's New entries for SAP Business Data Cloud, SAP Datasphere and SAP Analytics Cloud to the Business Data Cloud feed; the old BTP feed had stopped carrying them.Analytic modelling
SAP Business Data Cloudlinks revisedSAP publishes the What's New entries for SAP Business Data Cloud on its own feed; the entry linked none before.Cataloguing and lineage
SAP Dataspherelinks revisedSAP moved the What's New entries for SAP Business Data Cloud, SAP Datasphere and SAP Analytics Cloud to the Business Data Cloud feed; the old BTP feed had stopped carrying them.Analytic modelling
Tue 20
What an AI infra engineer sees when opening SAP's AI stack (1/3): the model gatewaypublishedA first look at SAP AI CoreSAP AI CoreSAP's AI service on BTP: the runtime for AI assets, and through the Generative AI Hub inside it the access to foundation models, the orchestration service with its templating, masking, filtering and grounding modules, the Prompt Registry and Document Grounding.sap official, btp's Generative AI Hub from someone who builds gateways outside SAP for a living: the orchestration service harmonizes provider calls onto one API (a model name string switches providers, and SAP Note 3437766 carries the current model list, token rates and deprecation dates), runs a fixed, non-reorderable module pipeline — templating, masking through SAP Data Privacy Integration, grounding, content filtering, then the model call — and retries down an ordered fallback list whose entries can each carry their own model and rules. It places @sap-ai-sdk/orchestration@sap-ai-sdk/orchestrationThe Orchestration Service client of SAP's AI SDK, which applies templating, grounding, filtering and masking around a model call rather than in the application.sap official, cap among the official JavaScript, Java and Python SDKs, and notes SAP's own LiteLLM provider as a way to put a different gateway in front of it.LLM integration
Mon 19
🟣 AI Eating Brains! How I Fed Mine to Joule Work DesktoppublishedDevtoberfest 2026 session AI-22. Mathias KemeterMathias Kemetergave 1 talk, contributed to 1 entry connects his Obsidian note vault to Joule WorkJoule WorkSAP’s workspace where people express a goal, work with the results in contextual Spaces and direct Joule Assistants and Joule Agents across SAP and third-party systems.sap official, umbrella Desktop with VaultGate, a local HTTP MCP server he wrote in his spare time because the desktop's sandbox accepts HTTP MCP servers and not stdio ones, and because he wanted the vault reached through Obsidian's own command line rather than by direct file access. The demo builds a Space of management tasks from notes and mailbox together, has the assistant pick up a conventions file in the vault without being told where the reporting rules are, finds a conflict the notes never name by semantic search over local embeddings, and appends a task to a sprint note behind a write confirmation, which is where the edges to Joule Work and Agent workspacesAgent workspacesGiving people a workspace to direct agents, retain the context of a task and interact with its results across connected applications. This is the user-facing experience around agents; building their logic belongs to Agent tooling, and controlling their calls belongs to Tool governance.SAP only hold. The second half is about the guardrails he had AI choose for the project it then had to live inside: tests and coverage, type safety, static checks, dependency and secret scanning, blocking on commit, on push and in CI.Agent workspaces
🟣 Bring AI into Your ABAP Applications with ISLM: Build, Integrate, and Consume AI NativelypublishedDevtoberfest 2026 session AI-11, from the product manager and product owner of Intelligent Scenario Lifecycle Management: ISLM as the place an AI use case is registered, deployed and operated inside an SAP S/4HANA system, and what has been added to it for generative AI. The demo builds a custom scenario end to end — an intelligent scenario of type generative AI, a connection to SAP AI CoreSAP AI CoreSAP's AI service on BTP: the runtime for AI assets, and through the Generative AI Hub inside it the access to foundation models, the orchestration service with its templating, masking, filtering and grounding modules, the Prompt Registry and Document Grounding.sap official, btp, a GPT-5 mini model, an orchestration flow template that configures data masking, input and output filtering and translation into the user's logged-in language, then prompts with a grounding query over five companies' return-policy documents — and consumes it from an ABAP report through the ABAP AI SDKABAP AI SDKSAP's reuse library for calling a large language model from ABAP code, powered by Intelligent Scenario Lifecycle Management and reaching the models in the Generative AI Hub; delivered with SAP BTP ABAP environment, SAP S/4HANA Cloud Public and Private Edition and SAP S/4HANA.sap official, abap, which calls the orchestration API and returns the summary as JSON. Two newer pieces follow: SAP RPT, the relational pre-trained transformer for tabular prediction, called with context rows and query rows and answering with a confidence value and no training step; and system-triggered agents, where an application job in the ABAP system starts a billing-creation agent that reaches S/4HANA OData APIs as tools through the agent gateway with principal propagation. The closing availability matrix states the generative AI capabilities for on-premise and private cloud from the 2021 release, with SAP notes to implement on lower ones.LLM integration
🟣 Enable Your AI application to Run on Multiple Models to Ensure Resilience and CompliancepublishedDevtoberfest 2026 session AI-21. A cross product owner on the SAP AI CoreSAP AI CoreSAP's AI service on BTP: the runtime for AI assets, and through the Generative AI Hub inside it the access to foundation models, the orchestration service with its templating, masking, filtering and grounding modules, the Prompt Registry and Document Grounding.sap official, btp team gives the reasons an AI application ends up on more than one model — a model down at the hyperscaler, a data centre that is not allowed to run it, a version being deprecated — and runs a joke app that answers from Gemini 2.5 Flash in EU10 and from Claude 4.5 Haiku in Asia-Pacific. The walk through AI Core's parts is where the edges hold: prompts kept as templates in the prompt registry, imported through the launchpad and the APIs or synced as YAML from a git repository (Prompt managementPrompt managementKeeping the prompts an application sends to a model as templates with placeholders, stored and versioned apart from the code that calls them, so a change of wording is not a deployment. What people conflate under it: the template filled at runtime, the registry it lives in, and the orchestration step that fills it.SAP only); the evaluation service scoring one prompt across up to four models per run on built-in or model-as-judge metrics, and the prompt optimizer generating variants against one metric (Agent evalsAgent evalsKnowing whether an AI feature still behaves after a model, prompt or schema change. Regression testing for non-deterministic output, against fixtures that reflect real business data.SAP and community, unassessed); and the orchestration service harmonising the providers behind one message format with a priority-sorted fallback list (LLM integrationLLM integrationCalling a language model from application code with the tenant, destination and credential handling already solved instead using an SDK call with a hand-rolled key.SAP and community, narrow by design). The session also states where a metric has to be designed rather than picked: an exact-JSON match scores an email generator at zero, so the joke app's custom metric judges topic and family-appropriateness instead.LLM integration
Achieve Agentic Enterprise Automation with SAP Integration Suite and Joule StudiopublishedA Devtoberfest 2026 integration-track session (INT-22), the second of that week's integration sessions, given by an SAP product expert for process automation and integration together with the product manager for Joule StudioJoule StudioSAP’s managed development environment for building custom agents, applications and workflows from business intent, in six phases from intent and specification to deployment on an SAP-managed runtime.sap official, umbrella, who walks the audience through it. The opening states the subject: process management, automation and orchestration unified into one combined approach, which the presenters put forward as the foundation for building custom AI agents.Agent tooling
Show 22 more
AI-assisted SAP development with MCPpublishedA Devtoberfest 2026 session (UI-41) that traces AI-assisted SAP development from pasting code into a chat window to coding agents with built-in connectors, and sorts the MCP servers around SAP development into SAP's own, the community's and the hosted ones, where the agent cannot change what the server allows. Four demos follow: GitHub Copilot alone explaining an ABAP report from a Git repository, @ui5/mcp-server@ui5/mcp-serverSAP's MCP server for UI5 development: ten tools that scaffold an app or a card, read the project in front of the agent, and run the linter and the manifest validator against it.sap official, ui5 reading an OpenUI5 sample app and converting it to TypeScript, mcp-sap-docsmcp-sap-docsAn MCP server over seventeen cloned SAP documentation repositories — ABAP keyword docs, capire, OpenUI5, Cloud SDK, BTP — searched with BM25 and prebuilt embeddings fused by reciprocal rank fusion.independent, cap grounding Terraform files for a BTP subaccount the presenter had not written before, and arc-1arc-1An MCP server in front of the ADT REST API that runs locally and on Cloud Foundry with twelve tools that read, search, write, activate, lint and transport ABAP objects.independent, abap with Copilot in Eclipse tracing why a title longer than 40 characters is rejected and proposing the DDIC-allowed length instead. It ends on a skill that migrates an SAP Gateway project on an ECC system to RAP on S/4HANA step by step, and on a rule Marian ZeisMarian Zeismaintains 35 entries, created 25 pieces, contributed to 86 entries gives for servers written for internal use: keep them read-only, because what the agent is short of is context.Agent access to your code
Be the First to Understand Joule Work's Agentic ThinkingpublishedA Devtoberfest 2026 session on the agent harness behind agentic thinking in Joule WorkJoule WorkSAP’s workspace where people express a goal, work with the results in contextual Spaces and direct Joule Assistants and Joule Agents across SAP and third-party systems.sap official, umbrella: generic operating-system tools in a gVisor-isolated sandbox instead of hand-wired domain functions, a react loop, a warm pool of sandboxes torn down after each task, tenant-separated skill loading, and placeholder tokens the agent never sees swapped for real credentials at the egress path. A skill is a folder with a SKILL.md carrying metadata and instructions, and where it reaches a system, an SAP connectivity JSON that declares each connector by ORD ID with its guardrails — GET allowed by default, POST, PUT, PATCH and DELETE requiring human approval, everything else blocked. The demo uploads a cost-centre skill through the skills view, then has agentic thinking delegate over A2A through Agent Gateway to a custom agent built in Joule StudioJoule StudioSAP’s managed development environment for building custom agents, applications and workflows from business intent, in six phases from intent and specification to deployment on an SAP-managed runtime.sap official, umbrella, whose MCP tools read cost centres from an S/4HANA Cloud system, and ends by turning the result into a PDF; the Joule Studio trace is what shows the tool call actually happened.Agent workspaces
Bridging Python and SAP HANA: Inside the sqlalchemy-hana Open-Source DialectpublishedA Devtoberfest 2026 data-track session (DAT-21) in which one of the maintainers of sqlalchemy-hana, the open-source SQLAlchemy dialect published in SAP's GitHub organisation, opens up what sits between Python and SAP HANASAP HANASAP’s relational database platform, which stores application data and executes SQL and SQLScript in the database, available on-premise and as the managed service SAP HANA Cloud.sap official, btp. It walks the five dialect classes that render statements, types, DDL, identifiers and sequences, and the HANA specifics they absorb: the FROM DUMMY a bare SELECT needs, a limit injected when only an offset is given, string compiled to NVARCHAR, and quoted case-sensitive identifiers against Python's lowercase. The rest covers schema migrations with Alembic on HANA's transactional DDL, the recently added async support and vector types, and that HANA Cloud and HANA 2 are supported while HANA 1 is not.Database connectors
Bring "developers.sap.com" Into Your Dev Environment: SAP Developer Context for AI Coding AgentspublishedDevtoberfest 2026 session FA-11: a tour of the rebuilt developers.sap.com by the team that runs it, from the administration screens behind the home page to the agent interfaces in front of it. The site is a CAP application on SAP HANA CloudSAP HANA CloudThe managed database under most of this: CAP's default target, the store Datasphere and Analytics Cloud sit on, and a service with plans and components.sap official, btp and SAP BTP, open-sourced in full, and the half of the session that places it here is how its content reaches an agent: every tutorial is also served as markdown, a coding agent is detected and given that version instead of the HTML, a well-known agent card advertises what the site offers, and SAP Developers MCPSAP Developers MCPThe hosted MCP server of developers.sap.com, over which an agent searches SAP's tutorials, reads missions and queries the site's knowledge graph of tutorials and concepts.sap official, umbrella answers questions about tutorials, news and a reader's own completion history in Claude Code and Joule Desktop, over OAuth or a personal access token. Behind the answers sit two things working together: a knowledge graph in the HANA database over 1400 tutorials and 11,000 concepts, and SAP AI CoreSAP AI CoreSAP's AI service on BTP: the runtime for AI assets, and through the Generative AI Hub inside it the access to foundation models, the orchestration service with its templating, masking, filtering and grounding modules, the Prompt Registry and Document Grounding.sap official, btp for the Joule interactions and the embeddings, with the agent harness written for the site rather than taken off a shelf. Where neither MCP nor A2A is available, the same content is published as skills.Agent guidance
Building Agentic Workflows with n8n in Joule StudiopublishedA Devtoberfest 2026 session, JS-31, on SAP-managed n8n inside Joule StudioJoule StudioSAP’s managed development environment for building custom agents, applications and workflows from business intent, in six phases from intent and specification to deployment on an SAP-managed runtime.sap official, umbrella, shown as a labs preview on a blocked-invoice scenario: a solution made of assets, an intent refined by the web agent into requirements and then a specification, and the generated workflow opened in the n8n editor that Joule Studio connects to its own n8n tenant, with the generated files committed through GitHub. The nodes carry the other edges — the human-in-the-loop node is built on SAP Task CenterSAP Task CenterOne inbox across SAP applications: tasks from S/4HANA, SuccessFactors, Ariba and BTP workflows in one list, with a public API for a custom task provider.sap official, btp, and SAP AI CoreSAP AI CoreSAP's AI service on BTP: the runtime for AI assets, and through the Generative AI Hub inside it the access to foundation models, the orchestration service with its templating, masking, filtering and grounding modules, the Prompt Registry and Document Grounding.sap official, btp is where model providers are configured centrally instead of per developer. It ends in Joule chat, where the approval task the workflow created is listed, opened and completed.Agent tooling
Building Code-Based AI Agents on the BAIPpublishedA Devtoberfest 2026 session on writing the agent yourself rather than configuring one, which is the agent toolingAgent toolingBuilding an agent into your own application — the tools it may call, the state it keeps between turns, and the framework it runs in. Distinct from MCP servers, which expose an application to somebody else's agent, and from agent guidance, which is knowledge handed to a coding agent.SAP and community, unassessed edge: an agent is a model with tools, skills, memory and context around it, and the session reads the code of one built in CAP as a LangGraph state graph, with an agent node that calls the model, a tools node holding the MCP clients, and the edge between them looping until no tool is needed. It then shows the shorter route, a CDS service annotated for the agent protocol with its instructions and skills in an agent.md file. SAP AI CoreSAP AI CoreSAP's AI service on BTP: the runtime for AI assets, and through the Generative AI Hub inside it the access to foundation models, the orchestration service with its templating, masking, filtering and grounding modules, the Prompt Registry and Document Grounding.sap official, btp carries both, with the CAP service in front of it as middleware so the credentials never reach the browser, and that is what cap-agent-ui5-webuicap-agent-ui5-webuiGives a CAP agent a UI5 web interface out of the box, so an agent built in the service has somewhere to be talked to without a separate frontend project.independent, cap is demonstrated against: added as a dependency, it gives the annotated service a UI5 chat interface that can be deployed rather than only previewed. A Python multi-agent application on Cloud Foundry, where an orchestrator delegates to per-stack specialists so no single model sees every tool, closes the session.Agent tooling
Building your knowledge graph on SAP HANA CloudpublishedDevtoberfest 2026 session DAT-11, in which two enterprise architects explain what a knowledge graph is and then build one on SAP HANA CloudSAP HANA CloudThe managed database under most of this: CAP's default target, the store Datasphere and Analytics Cloud sit on, and a service with plans and components.sap official, btp. The first half sets the join across three relational tables against an ontology of classes and predicates, the subject-predicate-object triple written as a turtle file, and a SPARQL query over it, and argues that the chain of relationships is what makes a recommendation traceable back to its reasons. The demo, taken from a Discovery Center mission, curates the ontology and the RDF files for an accounts receivable scenario in VS Code, ingests the triples into HANA Cloud with SQL and walks the resulting graph outward from a customer; that is the graph, spatial and document dataGraph, spatial and document dataStoring and querying data that is not plain rows, such as a network of relationships, a geometry or a JSON document, in the database that already holds the rows instead of in a separate store per model. Free-text search over entities is search; similarity over embeddings is retrieval-augmented generation.SAP only edge. It closes by separating three things that share the name: the HANA Cloud knowledge graph engine, SAP Knowledge Graph as the data, and the knowledge graph in SAP Business Data CloudSAP Business Data CloudSAP's umbrella over the data estate: installed intelligent content and data products, a governed catalog they are published through, and a path out to SAP Databricks.sap official, btp.Database connectors
Combining AI Agents with the power of SAP HANA embedded ML (part 1)publishedDevtoberfest 2026 session DAT-12, the first of two parts, in which Vitaliy RudnytskiyVitaliy Rudnytskiygave 1 talk, contributed to 6 entries does a data scientist's work against the embedded machine learning of SAP HANA CloudSAP HANA CloudThe managed database under most of this: CAP's default target, the store Datasphere and Analytics Cloud sit on, and a service with plans and components.sap official, btp by talking to an agent. The session first takes an agent apart into its LLM, its tools, its memory and the loop between them, then stacks the Predictive Analysis Library under the hana-ml Python wrapper and that under the HANA AI toolkit, whose tools an agent calls from plain English in a notebook to preview a sales refunds table, split it into train and predict tables, run a time-series check and train an AutoML forecast, keeping what it decided in Markdown memory files. That is the agent toolingAgent toolingBuilding an agent into your own application — the tools it may call, the state it keeps between turns, and the framework it runs in. Distinct from MCP servers, which expose an application to somebody else's agent, and from agent guidance, which is knowledge handed to a coding agent.SAP and community, unassessed edge; the same toolkit is then started as an MCP server over stdio and over streamable HTTP and reached from a coding agent and from Joule Work's desktop, which is where the live demo partly fails and the failure is shown.Database connectors
Data Product Governance in SAP Business Data CloudpublishedA Devtoberfest 2026 data-track session (DAT-41) on data product governance in SAP Business Data CloudSAP Business Data CloudSAP's umbrella over the data estate: installed intelligent content and data products, a governed catalog they are published through, and a path out to SAP Databricks.sap official, btp, released as an MVP: every data product belongs to a domain, a consumer puts one in a cart and raises an access request with a purpose, a target location and acknowledged terms and conditions, and a data steward's approval becomes an access agreement recording who uses which data product where. The demo runs both sides in the BDC cockpit — domains, access management and data product management for the steward, catalog, cart and request for the consumer, with Databricks as the target workspace — and that is what the Cataloguing and lineageCataloguing and lineageFinding what data exists across the estate and what it promises — catalogs and data products, the contract a consumer can rely on, where a field came from, and what a column actually contains once you profile it. The question asked before modelling starts and again when somebody asks where a number came from.SAP and community, unassessed edge rests on. It also states the current limits: one default domain until custom domains and subdomains arrive, auto-approved requests for space copies, transports and intelligent content, stewards who cannot yet revoke an active agreement, and an install step still done by hand after approval.Cataloguing and lineage
From Intent to Arcade: Building Joule*Bert with Joule Studio and AI-Assisted Pro CodepublishedA Devtoberfest 2026 AI-track session that takes a Q*bert-style arcade game, written as an Easter egg for colleagues, through every phase Joule StudioJoule StudioSAP’s managed development environment for building custom agents, applications and workflows from business intent, in six phases from intent and specification to deployment on an SAP-managed runtime.sap official, umbrella runs: intent analysis with its clarifying questions, a product requirements document, a specification with its task list, generated code, generated tests and deployment. The knowledge-graph lookup reports that the game maps to no enterprise process and the flow adjusts, and the presenter edits the requirements by hand when the first version declares sound out of scope. He then pushes the project to GitHub, clones it and initialises the jl command line@sap/joule-work-dev-cli-early-adoptersjl, the pro-code CLI of the new Joule Studio: a solution is a folder with a solution.yaml and typed assets, validated, built into a zip and deployed to Joule Studio's managed runtime, with Claude Code or OpenCode wired in over MCP. Early Adopter Care.sap official, btp so that Joule Studio's MCP servers are wired into the coding agent in VS Code, where mobile arrow keys were added afterwards; applications could not yet be deployed to the Joule Studio runtime, so this one runs on Kyma.Agent tooling
How to implement business AI capabilities in ABAP CloudpublishedA Devtoberfest 2026 session (ABA-32) on embedding AI in an ABAP Cloud application and on making SAP Joule aware of what that application can do. The first demo implements two features of a RAP travel app with the ABAP AI SDKABAP AI SDKSAP's reuse library for calling a large language model from ABAP code, powered by Intelligent Scenario Lifecycle Management and reaching the models in the Generative AI Hub; delivered with SAP BTP ABAP environment, SAP S/4HANA Cloud Public and Private Edition and SAP S/4HANA.sap official, abap: a determination on an uploaded receipt that sends the image together with the allowed categories and transaction types to a model reached through an intelligent scenario, then writes the extracted description, category, currency and amount back to the expense entity; and a recommendation function, triggered by a side effect on the trip title, that proposes a trip type with a score the Fiori UI sorts the proposals by and leaves the business user to accept — which is LLM integrationLLM integrationCalling a language model from application code with the tenant, destination and credential handling already solved instead using an SDK call with a hand-rolled key.SAP and community, narrow by design from ABAP code. The second demo builds a capability for SAP JouleSAP JouleSAP's unified AI engagement layer across SAP applications: Joule Work is where a user expresses intent, Joule Assistants coordinate agents across business functions, and Joule Agents carry out the tasks, provisioned on BTP as a formation against SAP Cloud Identity Services.sap official, btp in the Joule Studio code editor in VS Code, with a scenario that extracts a trip title from the user's intent and calls an OData service in the ABAP backend through a BTP destination, deploys it and asks the Joule test UI for the trip's details; the session states that agents reach ABAP only over governed remote APIs — OData, SQL and MCP — and that automatic grounding of custom OData services in the knowledge graph is upcoming rather than available.LLM integration
How We Make Tabular Predictions in 2026 with TabPFNpublishedA Devtoberfest 2026 session, AI-32, on why a table needs a model of its own: a language model reasons its way to a plausible number, while a tabular foundation model takes the table as context in one forward pass and learns from its features statistically. That is predictive modelsPredictive modelsTraining, deploying and calling a model that predicts a value from tabular business data: a classification, a regression, a forecast. Distinct from LLM integration, where a language model generates text, even where both run on the same service.SAP only from both ends — a scikit-learn-shaped fit and predict in a notebook over a heart-disease data set, predicted class probabilities and SHAP values for which features drove the outcome, then the same model reached as an MCP tool by a coding agent that is asked which current customers are most likely to churn. The presenter is from the lab that publishes TabPFN; the Q&A puts it beside SAP's own tabular foundation model, which she says she has not used.Predictive models
Meet Joule Studio and Build your First AI AgentpublishedA Devtoberfest 2026 session from the product manager of Joule StudioJoule StudioSAP’s managed development environment for building custom agents, applications and workflows from business intent, in six phases from intent and specification to deployment on an SAP-managed runtime.sap official, umbrella, walking one agent through the whole path the tool defines: an intent, the problem statement and goals a coding agent derives from it, the business context it takes from the SAP knowledge graph, and the solution type it proposes with a stated intent fit. The demo is a collections agent that drafts dunning emails: the product requirements document names the metrics and the human-in-the-loop review, the technical specification lists the artifacts to build — the MCP tools among them — and the build is followed by a business test run against business partner 10001, technical unit tests, and deployment to the managed runtime, where the agent and its MCP server appear as deployed assets. Two points shape how it is used rather than what it produces: every asset is a file, editable in the web editor or in the code view and pushable to git, and the same solution can be built from local pro-code tools; n8n workflows are stated as delivered out of the box inside the studio.Agent tooling
Real Pro-Code Development with Joule StudiopublishedA Devtoberfest 2026 AI-track session that runs one agent build across both of Joule StudioJoule StudioSAP’s managed development environment for building custom agents, applications and workflows from business intent, in six phases from intent and specification to deployment on an SAP-managed runtime.sap official, umbrella's paths. Playing a procurement manager, one presenter states the intent in the browser, answers the clarifying questions, watches the fit-gap analysis name the S/4HANA APIs the use case needs, and pushes the intent and requirements document to GitHub; the other clones it and carries on in a terminal, logging in with the jl command line@sap/joule-work-dev-cli-early-adoptersjl, the pro-code CLI of the new Joule Studio: a solution is a folder with a solution.yaml and typed assets, validated, built into a zip and deployed to Joule Studio's managed runtime, with Claude Code or OpenCode wired in over MCP. Early Adopter Care.sap official, btp against her landscape URL and running its init command to fetch Joule Studio's skills for OpenCode. From there the coding agent writes the specification, the discovered API specifications, the MCP translation and the agent's Python file with its model configuration and evaluation scenarios, which is the edge to Agent toolingAgent toolingBuilding an agent into your own application — the tools it may call, the state it keeps between turns, and the framework it runs in. Distinct from MCP servers, which expose an application to somebody else's agent, and from agent guidance, which is knowledge handed to a coding agent.SAP and community, unassessed; the skills themselves, which SAP says not to edit but which a developer may add to, are the edge to Agent guidanceAgent guidancePackaged knowledge you hand a coding agent so it gets a framework right unprompted — skills, rule files, instruction packs. Distinct from agent tooling, which is about building agentic features into your application, and from agent access to your code, where the agent reads the real project instead of being handed a file about it.SAP and community, core adopted it.Agent tooling
SAP BDC and AI: A Perfect MatchpublishedA Devtoberfest 2026 data-track session (DAT-22) in which Business Data Cloud's CTO sets out how SAP Business Data CloudSAP Business Data CloudSAP's umbrella over the data estate: installed intelligent content and data products, a governed catalog they are published through, and a path out to SAP Databricks.sap official, btp feeds SAP's business AI platform: data products whose primary persistence is an object store and whose semantics are kept with them, the one domain model that links a supplier in Ariba to a supplier in S/4HANA, and the SAP knowledge graph, a triple store running in SAP HANA Cloud and queried with SPARQL, that SAP JouleSAP JouleSAP's unified AI engagement layer across SAP applications: Joule Work is where a user expresses intent, Joule Assistants coordinate agents across business functions, and Joule Agents carry out the tasks, provisioned on BTP as a formation against SAP Cloud Identity Services.sap official, btp consults before answering. The live demo asks Joule for the available data products and the top ten open purchase orders, with the data product query agent compiling SQL over virtual tables and a space rendering a dashboard, which is the cataloguing side of this: a central registry, the Unified Customer Landscape, that every activated data product is registered in. The Q&A covers where the knowledge graph lives and what a customer may extend, base policies that turn into filter conditions on an agent's queries, and Joule StudioJoule StudioSAP’s managed development environment for building custom agents, applications and workflows from business intent, in six phases from intent and specification to deployment on an SAP-managed runtime.sap official, umbrella as the place to build your own agents and data products.Cataloguing and lineage
SAP BDC in Business AI Platform: Contextualize and ReasonpublishedA Devtoberfest 2026 session (DAT-31) that places SAP Business Data CloudSAP Business Data CloudSAP's umbrella over the data estate: installed intelligent content and data products, a governed catalog they are published through, and a path out to SAP Databricks.sap official, btp as the contextualise-and-reason pillar of the SAP Business AI Platform, between the build pillar of Joule Studio and Integration Suite and the govern pillar of AI Agent Hub. It covers data products described with open resource discovery and core schema notation metadata, a data product creation agent that assembles a derived data product for overdue billing documents from the billing document and customer products and shows its lineage and fields; the data products a line-of-business application brings being readable by Joule without a BDC subscription; and the SAP-managed knowledge graph that maps business entities and their relationships for agents, held as graph dataGraph, spatial and document dataStoring and querying data that is not plain rows, such as a network of relationships, a geometry or a JSON document, in the database that already holds the rows instead of in a separate store per model. Free-text search over entities is search; similarity over embeddings is retrieval-augmented generation.SAP only in the SAP HANA CloudSAP HANA CloudThe managed database under most of this: CAP's default target, the store Datasphere and Analytics Cloud sit on, and a service with plans and components.sap official, btp knowledge graph engine as RDF triples queried with SPARQL. The closing demo has SAP JouleSAP JouleSAP's unified AI engagement layer across SAP applications: Joule Work is where a user expresses intent, Joule Assistants coordinate agents across business functions, and Joule Agents carry out the tasks, provisioned on BTP as a formation against SAP Cloud Identity Services.sap official, btp answer head-count and purchase-order questions, with the generated SQL passed through a data product query that adds the restriction policy defined for that user in cloud identity services, which is why the session stands for cataloguing and lineageCataloguing and lineageFinding what data exists across the estate and what it promises — catalogs and data products, the contract a consumer can rely on, where a field came from, and what a column actually contains once you profile it. The question asked before modelling starts and again when somebody asks where a number came from.SAP and community, unassessed as much as for the products it names.Cataloguing and lineage
SAP HANA Data Lake Files (HDLF) Basic FunctionalitiespublishedA Devtoberfest 2026 data-track session on SAP HANA data lake files, the managed object store inside SAP HANA CloudSAP HANA CloudThe managed database under most of this: CAP's default target, the store Datasphere and Analytics Cloud sit on, and a service with plans and components.sap official, btp that is provisioned from HANA Cloud Central, priced per API call and per stored gigabyte, and reached with the same REST API whichever hyperscaler it runs on. Two product managers set up an instance with its IP allow list, client certificates and trusts, then upload, list, read and delete files over the WebHDFS REST API and the HDLFS command-line utility — the Object storageObject storageStoring application files as objects in a bucket or container backed by a cloud storage service. Connecting a file to a business record is the separate Attachments concern.SAP only half — and cover file versions with snapshots, access policies per file, and chunked upload with a server-side merge. They then link the instance to a HANA database and query a CSV file as a virtual table: SQL on files now runs through a file adapter in the database, while the older path through the data lake relational engine is deprecated.Database connectors
Strategy and Tool Updates on BW Modernization with SAP Business Data CloudpublishedDevtoberfest 2026 session DAT-13, in which a product manager for SAP Business Data CloudSAP Business Data CloudSAP's umbrella over the data estate: installed intelligent content and data products, a governed catalog they are published through, and a path out to SAP Databricks.sap official, btp sets out the revised BW modernisation strategy, now called optimise, modernise and innovate, and the three tools that carry it. The data product generator transfers BW InfoProviders into local table files in a dedicated BW inbound space of SAP DatasphereSAP DatasphereSAP's cloud data warehouse: spaces, modelled views and analytic models over data brought in or federated from source systems, with the business context carried along rather than rebuilt.sap official, btp, where a data product is still created by hand in the data sharing cockpit; the query template generator turns BW query metadata into an analytical model template in the HANA space; the BW migration assistant, planned to start in Q4 for BW NetWeaver, takes the data flows and transformations below them, and is shown on slides only. That is the loading a warehouseLoading a warehouseGetting data into an analytical store and keeping it current — connections to source systems, replication and the other integration tasks, import jobs and their lifecycle, and the monitoring that tells you a load failed before a report does. Distinct from replication, which is an application reading data that lives elsewhere as if it were local rather than a warehouse being filled.SAP and community, unassessed edge, and the demos cover the BW/4HANA cockpit, the ABAP program for the NetWeaver GUI, mass generation with semantics from the BW modeling tools, and the flattening operator for sharing data outside SAP.Cataloguing and lineage
Supercharge Joule Desktop: Connect Any SAP API in Minutes with MCP GatewaypublishedDevtoberfest 2026 session INT-13, in which an SAP solution advisor connects a live S/4HANA system to the desktop experience of Joule WorkJoule WorkSAP’s workspace where people express a goal, work with the results in contextual Spaces and direct Joule Assistants and Joule Agents across SAP and third-party systems.sap official, umbrella through the MCP gateway in SAP Integration SuiteSAP Integration SuiteSAP's integration service on BTP: Cloud Integration for orchestration, API Management in front of it, and an MCP gateway that serves a managed API to agents.sap official, btp. One OpenAPI specification covering business partners, sales orders, material stock, billing and purchase requisitions is uploaded as an MCP server artifact, deployed to integration cell with XSUAA inbound authentication, and registered as an MCP connector; the destination behind it is then switched from basic authentication to principal propagation, and the Cloud ConnectorCloud ConnectorSAP's on-premises agent that links SAP BTP applications to systems in a customer's own network, as a reverse invoke proxy that exposes only the systems and resources it is configured to allow.sap official, btp monitor shows the same question arriving as the presenter's own user. That demonstration is the serve your application over MCPServe your application over MCPExposing selected application data and business operations to an AI agent through the Model Context Protocol. An application may also integrate remote services through service integration, connecting its exposed contract with agent access to your systems. Development assistance belongs to agent access to your code.SAP and community, narrow by design edge, and the second half is Joule Work itself: a Markdown skill that fixes the steps and format of a customer briefing, and a Space the skill rebuilds with each run.Agent workspaces
Unlocking Tabular Foundation Models: From Experimentation to ProductionpublishedA Devtoberfest 2026 session (AI-41) on taking tabular prediction from a playground to a deployment. It sets tabular foundation models against a model trained per use case and against language models reading rows as text, explains in-context learning as examples plus the rows whose target column is missing, and tries SAP RPT 1.6 on a train-delay dataset in the RPT playground, including the column and row importance it returns with the prediction. The same prediction is then rebuilt on SAP AI CoreSAP AI CoreSAP's AI service on BTP: the runtime for AI assets, and through the Generative AI Hub inside it the access to foundation models, the orchestration service with its templating, masking, filtering and grounding modules, the Prompt Registry and Document Grounding.sap official, btp, where the tabular orchestration service plays the part the orchestration service plays for language models: a data destination, a table artifact with its CSN schema and a scenario configuration registered once at design time, after which a predict call passes only the query rows, the scenario configuration and a context-selection strategy. The last demo runs those same calls in natural language through the aicore-skillsaicore-skillsFour agent skills for SAP AI Core, published as a plugin inside SAP's AI Skills Library: admin resources, the deployment lifecycle, the foundation models in the Generative AI Hub, and the Context Registry behind tabular prediction.sap official, btp plugin in a coding agent, and one of them only resolves once the agent is told which scenario configuration to use.LLM integration
Agent workspacesecosystems revisedThe ABAP, CAP and UI5 views existed only to hold Werkbank, whose stack-specific part is its catalog of development skills and connectors, not the workspace. That contribution moved to Agent guidance; nothing behind those doors offers a workspace of its own.Agent workspaces
secondsky/sap-skillstype properties, flavors revisedRead today in the two plugins that answer the CAP and the UI5 door, which the ABAP pass had not opened. The sap-cap-capire skill's frontmatter states that it covers Node.js and Java runtimes, so CAP's two flavors rest on the maintainer's sentence where the entry previously answered in both by the serves-another-door fallback, which is the same reading off an absence. The sapui5 skill's frontmatter names freestyle apps and Fiori elements apps in one clause, so freestyle is stated, and fiori-elements moves onto that sentence from the README line about which request activates sap-fiori-tools, which said what triggers a plugin rather than what it covers. Nothing read names UI5 Web Components, so that flavor stays unstated. Stating freestyle changes an answer: the entry answered no there while it stated one UI5 flavor and now answers yes, which a reader comparing that flavor's options will see. The ABAP pair is unchanged.Agent guidance