onedevatlasBeta
Browse the atlas
AllCAPUI5ABAPBTP
CDS pluginCDS plugintypeInstalls into a CAP project and extends the runtime. CAP loads it through cds.requires or the plugin hook.See every one →What this type is →independently maintainedindependently maintainedoriginPublished by one person or a small company outside SAP. Says nothing about quality; the readings say that.See every one →

cap-llm-plugin

Generative AI in a CAP application with the sensitive fields anonymised through SAP HANA before the prompt leaves, so the model never sees the confidential half.

npm58/wk

Facts

Latestv1.5.1ReleasedLicenceApache License 2.0, January 2004

Where it sits

Lives in
CAP
Provides
LLM integrationCAPLLM integrationcapabilityCalling 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.Data maskingCAPData maskingcapabilityReplacing the personal or confidential values in a prompt with placeholders before it leaves for a model, and putting them back in the answer. Anonymisation, where the values are gone for good, and pseudonymisation, where they return, are the two forms; which fields count as sensitive is the application's question, not the module's.Retrieval-augmented generationCAPRetrieval-augmented generationcapabilityGrounding a model's answer in data retrieved at question time: text split into chunks, each chunk embedded as a vector, and the chunks nearest to the question placed into the prompt. Semantic search is the same retrieval with a person reading the matches, and fine-tuning changes the model's weights, not its input.
Flavors
Node.jsfrom its type
Requires
@sap/cds≥ 7.1.1@sap/cdsartifactThe peer range, as published.
Optional
SAP HANA CloudSAP HANA CloudartifactThe vector engine the similarity search runs in, and the database the anonymisation happens in before a prompt leaves.SAP AI CoreSAP AI CoreartifactThe embedding and chat models the plugin calls, and the service the retrieval method reaches them through.
Designs
Generative AI on SAP BTPGenerative AI on SAP BTPreference architectureplugin within the application logic for Generative AI solutions

Before choosing this

Conditions

LicenceSAP Developer License Agreement 3.2 CAP§ 1, 2, 3A cds-plugin is used with SAP CAP materials, or an artifact is explicitly identified as a CAP plugin or application.
  • § 1 License. Productive customer applications are limited to SAP BTP or another SAP-licensed platform, and mass extraction to a non-SAP product is restricted except where solely required for SAP interoperability.
  • § 2 Intellectual Property. The agreement retains SAP's rights in the covered materials and limits making APIs, tools or software available to third parties, while preserving ownership of customer applications subject to SAP's underlying rights.
  • § 3 Artificial Intelligence Training. The agreement prohibits use of the covered materials and customer applications for training artificial-intelligence models or systems.
SAP Developer License Agreement 3.2 CAP →

Also, from LLM integration§ 3: Calling a model for inference is not automatically AI training. Section 3 matters when the covered material or application participates in training, model development, refinement or the text-and-data-mining uses it names.

Also, from Retrieval-augmented generation§ 3: Retrieval and inference are not automatically AI training. The section becomes relevant if the indexed or retrieved covered material is instead used for training, developing or refining the model or system.

PracticalFunctional boundary

Not where the records sit anywhere but SAP HANA Cloud: the similarity search and the single retrieval method are both written against its vector engine, and the embeddings and the chat call go through SAP AI Core, so a project without both has nothing here to call. Not where the dependency has to be an SAP-supported one either: the README states that the plugin is not an official SAP product and is published as a community contribution.

What it declares

Every artifact

Type
CDS plugin
Origin
Independently maintained

As a CDS plugin

Declared for
>=7.1.1
Distributed through
does not say

For LLM integration

Models it reaches
SAP AI Core

Tasks this contributes to2Beta

RAG on CDS1
Answer a question from records in a CAP serviceA question asked in plain words is answered from rows the CAP model already holds: the records nearest to the question in meaning are found and put in front of a language model, so the answer rests on those records rather than on what the model remembers.Covers, under conditionsretrieval in one call
What it does

Collapses the steps into one method: the README's retrieval feature embeds through SAP AI Core, searches SAP HANA Cloud's vector engine and hands the result to the model, with the embedding and the similarity search also available on their own for a handler that wants them apart.

Conditions

The vectors are in SAP HANA Cloud's vector engine, which every retrieval feature in the README is written against; An SAP AI Core binding provides the embedding and chat models

Remaining work

Fill the vector table first: the README describes generating embeddings and searching them, and neither chunking nor an embed-on-write handler is part of it. The README also states that the plugin is not an official SAP product and is published as a community contribution.

All options on the RAG on CDS page →
LLM integration1
Change which model a CAP service callsA CAP service that already calls a language model is pointed at a different one — a newer version, a cheaper one, or a model from another vendor — and the handler that calls it is not rewritten.Also in BTP/BAIP, ABAPCovers, under conditionsplugin with an orchestration route
What it does

Reaches the same harmonised API from a CAP service: the README's feature list names support for SAP AI Core's orchestration service harmonised chat-completion APIs, so a service already calling the plugin can take a different model through that route.

Conditions

The orchestration route is used: the README lists support for the orchestration service's harmonised chat-completion APIs beside its own simplified chat-completion method, and only the harmonised API is what SAP describes as letting the model change without the client code changing; The records and the models stay on SAP HANA Cloud and SAP AI Core, which every feature in this README is written against

Remaining work

Read which of the plugin's two chat paths an existing handler uses, because the simplified chat-completion method is not the harmonised one; the README does not say what changes in the handler when the model does. The README also states that the plugin is not an official SAP product and is published as a community contribution.

All options on the LLM integration page →

Connections5

Download
cap-llm-plugin connectionscap-llm-pluginPROVIDES→ provides3provides stated Anonymises the sensitive fields in SAP HANA Cloud before the prompt leaves for the model, and puts the genuine values back into the response, so the masking happens in the database rather than in the orchestration call. cap-llm-plugin — npm, README: https://www.npmjs.com/package/cap-llm-plugin cap-llm-plugin — npm, README: https://www.npmjs.com/package/cap-llm-pluginData maskingprovides stated Puts the chat call behind one method on a CAP service: the README's feature list names simple access to models through [SAP AI Core](artifact:btp-ai-core) with a simplified chat-completion method, and support for the orchestration service's harmonised chat-completion APIs beside it. Which model answers is configured in AI Core rather than chosen here. cap-llm-plugin 1.5.1 — README on unpkg: https://unpkg.com/cap-llm-plugin@1.5.1/README.md cap-llm-plugin 1.5.1 — README on unpkg, orchestration: https://unpkg.com/cap-llm-plugin@1.5.1/README.md cap-llm-plugin 1.5.1 — README on unpkg, not an SAP product: https://unpkg.com/cap-llm-plugin@1.5.1/README.mdLLM integrationprovides stated Puts the retrieval into one call: a single method that embeds the question through [SAP AI Core](artifact:btp-ai-core), searches [SAP HANA Cloud](artifact:btp-hana-cloud)'s vector engine for the nearest rows and sends them on to the model. The embedding and the similarity search are also exposed on their own, for a handler that wants to keep the steps apart. What gets embedded, and when, is the application's: the README describes no chunking and no embed-on-write. cap-llm-plugin 1.5.1 — README on unpkg: https://unpkg.com/cap-llm-plugin@1.5.1/README.md cap-llm-plugin 1.5.1 — README on unpkg: https://unpkg.com/cap-llm-plugin@1.5.1/README.md cap-llm-plugin 1.5.1 — README on unpkg: https://unpkg.com/cap-llm-plugin@1.5.1/README.mdRetrieval-augmented generationNEEDS→ optionally needs2optionally needs stated If: An embedding, a chat completion or the single retrieval method is called; each of them goes through SAP AI Core. The embedding and chat models the plugin calls, and the service the retrieval method reaches them through. cap-llm-plugin 1.5.1 — README on unpkg: https://unpkg.com/cap-llm-plugin@1.5.1/README.md cap-llm-plugin 1.5.1 — README on unpkg: https://unpkg.com/cap-llm-plugin@1.5.1/README.mdSAP AI Coreifoptionally needs stated If: The similarity search, the single retrieval method or the anonymisation are used; all three are written against SAP HANA Cloud. The vector engine the similarity search runs in, and the database the anonymisation happens in before a prompt leaves. cap-llm-plugin 1.5.1 — README on unpkg: https://unpkg.com/cap-llm-plugin@1.5.1/README.mdSAP HANA Cloudif

Install

npm add cap-llm-plugin

Derived from the package name. No recipe recorded; the README has the configuration.

Learn more3

Readings
  1. HANA Cloud’s VECTOR EMBEDDING in CAP and Comparison with OpenAI Embeddinglalitmohan, SAP Community
  2. RAG with SAP HANA Cloud Vector Engine, GenAI Hub & CAPKevinR, SAP Community
  3. Elevating Data Privacy in SAP CAP based GenAI Applications with CAP LLM PluginAkashAmarendra, SAP Community

Where it sits in your app

Download
Your CAP applicationYour SAP BTP subaccountdocumentedmay needif The similarity search, thesingle retrieval method or theanonymisation are used; all thre…documentedmay needif An embedding, a chatcompletion or the singleretrieval method is called; each…cap-llm-plugin, modulecap-llm-pluginSAP HANA Cloud, service instanceSAP HANA CloudSAP AI Core, service instanceSAP AI Core
SAPnot SAP's, or the project's ownunder a condition

README

quoted from cap-llm-plugin@1.5.1, read
cap-llm-plugin@1.5.1/README.mdOn npm

CAP LLM Plugin

CAP LLM Plugin helps developers create tailored Generative AI based CAP applications:

  1. Without exposing confidential data to LLM by anonymizing sensitive data leveraging SAP HANA Cloud Data Anonymization.
  2. Seamlessly generate vector embeddings via SAP AI Core.
  3. Easily retrieve Chat Completion response via SAP AI Core.
  4. Efforlessly perform similarity search via SAP HANA Cloud Vector engine.
  5. Simplified single RAG (retrieval-augmented generation) retrieval method powered by SAP AI Core and SAP HANA Cloud Vector Engine.
  6. Access the harmonized chat completion API of the SAP AI Core Orchestration service.
excerpt, Apache License 2.0, January 2004, read licencethe rest, on npm

History

Latest

v1.5.1

  1. v1.5.0

  2. v1.4.5

  3. v1.4.4

  4. v1.4.2

  5. v1.3.3

  6. v1.3.2

  7. v1.3.1

  8. v1.3.0

  9. v1.2.0

On npm
56downloads per week, npm
Weekly downloads for cap-llm-plugin26 complete calendar weeks from 2026-04-20 to 2026-10-18; the latest total is 56.20 Apr – 26 Apr: 22327 Apr – 3 May: 2144 May – 10 May: 37211 May – 17 May: 30918 May – 24 May: 15625 May – 31 May: 2121 Jun – 7 Jun: 1928 Jun – 14 Jun: 13415 Jun – 21 Jun: 20722 Jun – 28 Jun: 24929 Jun – 5 Jul: 1656 Jul – 12 Jul: 23513 Jul – 19 Jul: 8820 Jul – 26 Jul: 9927 Jul – 2 Aug: 1253 Aug – 9 Aug: 9110 Aug – 16 Aug: 49117 Aug – 23 Aug: 5424 Aug – 30 Aug: 7531 Aug – 6 Sept: 887 Sept – 13 Sept: 8514 Sept – 20 Sept: 6021 Sept – 27 Sept: 14228 Sept – 4 Oct: 1035 Oct – 11 Oct: 6212 Oct – 18 Oct: 56

Others providing18

LLM integration
  1. SAP AI CoreAI runtime on BTP with the Generative AI Hub
  2. cap-java/cds-aiBridge from CAP Java to SAP AI Core, with Fiori field recommendations
  3. @cap-js/aiValue-help recommendations and SAP AI Core as a CAP service
  4. @geertjanklaps/cds-toon-ai-protocol-adapterProtocol adapter serving CAP data as TOON instead of JSON
  5. @jerome-benoit/sap-ai-providerVercel AI SDK provider for SAP AI Core
  6. ui5-cap-ai-featuresCDS plugin and UI5 controls for AI text features via SAP AI Core
  7. @sap-ai-sdk/ai-apiAI Core control plane API surface of SAP's AI SDK
  8. @sap-ai-sdk/coreShared internals of SAP's AI SDK
  9. @sap-ai-sdk/foundation-modelsClient for foundation models via SAP AI Core and Generative AI Hub
  10. @sap-ai-sdk/orchestrationOrchestration Service client for templating, grounding and masking
  11. @sap-ai-sdk/prompt-registryClient for prompts kept in SAP's prompt registry
  12. @saptarishi/cds-plugin-llmCAP service kind putting eleven model providers behind one API
  13. @shahid.la/cds-db-nlquery-mcpMCP server answering plain-English questions against a CDS schema
Data masking
  1. SAP AI CoreAI runtime on BTP with the Generative AI Hub
  2. @sap-ai-sdk/orchestrationOrchestration Service client for templating, grounding and masking
RAG on CDS
  1. SAP AI CoreAI runtime on BTP with the Generative AI Hub
  2. SAP HANA CloudManaged SAP HANA database service on BTP
  3. @cap-js/aiValue-help recommendations and SAP AI Core as a CAP service
  4. @mi8y/cap-agents-aicore-vectorstoreLangChain vector store on SAP AI Core Document Grounding
  5. @mi8y/cap-agents-cds-vectorstoreLangChain vector store with CDS entities for embeddings
  6. @sap/cdsCAP Node.js runtime
  7. @saptarishi/cds-plugin-llmCAP service kind putting eleven model providers behind one API
  8. @saptarishi/cds-plugin-vector-hanaVector storage and similarity search on SAP HANA from CAP

Details

Sources4
npm registrydiscovered herechecked
The discovery passdiscovered herechecked
npm registry, cap-llm-pluginversion, releases, installschecked
README.mdquoted, and what the entry declareschecked

Sources are graph entities: the same page may back several entries, and an unreachable source flags every row that rests on it.

Compatibility and provenance
declared for>=7.1.1
latestv1.5.1
first release
releases and cadence10, ~63d median
discovered vianpm registry, The discovery pass
first seen in the atlas
ecosystem and domaincap, ai-and-agents
History in the atlas8
  1. an edge with what_it_does, three sources and the model_access property
  2. capability:llm-integration, capability:data-masking, capability:retrieval-augmented-generation
  3. Not where the records sit anywhere but SAP HANA Cloud, and not where the dependency has to be an SAP-supported one.
  4. artifact:btp-hana-cloud, artifact:btp-ai-core
  5. capability:llm-integration, capability:data-masking
  6. npm: https://www.npmjs.com/package/cap-llm-plugin
  7. Generative AI in a CAP application with the sensitive fields anonymised through SAP HANA before the prompt leaves, so the model never sees the confidential half.
  8. First seen.
For your agent
Ref
artifact:cap-llm-plugin
JSON
/artifacts/cap-llm-plugin.json
MCP
Connecting an agent