Sap Cloud SDK AI

SAP Cloud SDK for AI

Related Skills

• sap-cap-capire: Use for building CAP applications that consume AI services, including event handler patterns and async LLM orchestration
• sap-ai-core: Use for AI Core platform setup, orchestration configuration, and model deployment
• sap-dependency-security: Hardening guidance for npm package upgrades, lockfile policies, and secure dependency workflows

The official SDK for SAP AI Core, SAP Generative AI Hub, and Orchestration Service. Package versions are verified against public registries; AI Core tenant execution and exact model availability still require target-tenant validation.

When to Use This Skill

Use this skill when:

• Integrating AI/LLM capabilities into SAP BTP applications
• Building chat completion or embedding features
• Using tenant-approved OpenAI, Claude, Gemini, Amazon, Mistral, or other model families via SAP AI Core
• Implementing content filtering, data masking, or document grounding
• Creating agentic workflows with LangChain or Spring AI
• Managing prompts via Prompt Registry
• Deploying AI models on SAP AI Core

Table of Contents

• Quick Start
• Prerequisites
• Connection Setup
• Available Packages
• Supported Models
• Core Features
• Bundled Resources

Quick Start

Note: This skill uses SAP Cloud SDK for AI JavaScript v2.11.0+ and Java v1.19.0+ based on public package registry evidence from 2026-06-15. If you're migrating from v1.x, see V1 to V2 Migration Guide for breaking changes.

JavaScript/TypeScript

npm install @sap-ai-sdk/orchestration@^2

import { OrchestrationClient } from '@sap-ai-sdk/orchestration';

const client = new OrchestrationClient({
  promptTemplating: {
    model: { name: 'gpt-4o' },
    prompt: [{ role: 'user', content: '{{?question}}' }]
  }
});

const response = await client.chatCompletion({
  placeholderValues: { question: 'What is SAP?' }
});
console.log(response.getContent());

Java

<dependency>
  <groupId>com.sap.ai.sdk</groupId>
  <artifactId>orchestration</artifactId>
  <version>${ai-sdk.version}</version>
</dependency>

var client = new OrchestrationClient();
var config = new OrchestrationModuleConfig()
    .withLlmConfig(OrchestrationAiModel.GPT_4O);
var prompt = new OrchestrationPrompt("What is SAP?");
var result = client.chatCompletion(prompt, config);
System.out.println(result.getContent());

Prerequisites

• Node.js 20+ (JavaScript) or Java 17+ (Java)
• SAP AI Core service instance (extended or sap-internal plan)
• Orchestration deployment in AI Core (default resource group has this)

Connection Setup

BTP Runtime (Cloud Foundry/Kyma)

Bind AI Core service instance to your application. SDK auto-detects via VCAP_SERVICES or mounted secrets.

Local Development

Set environment variable:

export AICORE_SERVICE_KEY='{"clientid":"...","clientsecret":"...","url":"...","serviceurls":{"AI_API_URL":"..."}}'

Or use CAP hybrid mode:

# JavaScript
cds bind -2 <AICORE_INSTANCE> && cds-tsx watch --profile hybrid

# Java
cds bind --to aicore --exec mvn spring-boot:run

For detailed connection options, see references/connecting-to-ai-core.md

Available Packages

JavaScript/TypeScript

Java

Supported Models

Model IDs and versions are tenant-specific. Before copying an example into application code, list the target catalog through SAP AI Launchpad Model Library or the AI Core model-list API.

Example Families

• OpenAI: GPT-family chat, multimodal, reasoning, and embedding models where entitled
• Anthropic (AWS): Claude-family models where entitled
• Amazon: Nova/Titan-family models where entitled
• Google: Gemini-family models where entitled
• Mistral: Mistral-family models where entitled

Deprecated Models (Use Replacements)

Core Features

Chat Completion with Streaming

// JavaScript
const stream = client.stream({
  placeholderValues: { question: 'Explain SAP CAP' }
});

for await (const chunk of stream.toContentStream()) {
  process.stdout.write(chunk);
}

// Java
client.streamChatCompletion(prompt, config)
    .forEach(chunk -> System.out.print(chunk.getDeltaContent()));

Function/Tool Calling

// JavaScript
const tools = [{
  type: 'function',
  function: {
    name: 'get_weather',
    parameters: { type: 'object', properties: { city: { type: 'string' } } }
  }
}];

const response = await client.chatCompletion({
  placeholderValues: { question: 'Weather in Berlin?' }
}, { tools });

const toolCalls = response.getToolCalls();

Content Filtering

// JavaScript
import { buildAzureContentSafetyFilter } from '@sap-ai-sdk/orchestration';

const client = new OrchestrationClient({
  promptTemplating: { model: { name: 'gpt-4o' } },
  filtering: {
    input: buildAzureContentSafetyFilter({ Hate: 'ALLOW_SAFE' }),
    output: buildAzureContentSafetyFilter({ Violence: 'ALLOW_SAFE' })
  }
});

Data Masking

// JavaScript
const client = new OrchestrationClient({
  promptTemplating: { model: { name: 'gpt-4o' } },
  masking: {
    masking_providers: [{
      type: 'sap_data_privacy_integration',
      method: 'anonymization',
      entities: [{ type: 'profile-email' }, { type: 'profile-person' }]
    }]
  }
});

Document Grounding

// JavaScript
const client = new OrchestrationClient({
  promptTemplating: { model: { name: 'gpt-4o' } },
  grounding: {
    grounding_input: ['{{?question}}'],
    grounding_output: ['{{?context}}'],
    data_repositories: [{ type: 'vector', id: 'my-repo-id' }]
  }
});

CAP Integration

The SDK integrates natively with CAP event handlers. Use OrchestrationClient inside CAP service classes to add AI capabilities to your CAP services.

Service binding in MTA:

resources:
  - name: my-ai-core
    type: org.cloudfoundry.managed-service
    parameters:
      service: aicore
      service-plan: extended

CAP event handler with AI:

import { OrchestrationClient } from '@sap-ai-sdk/orchestration';
import cds from '@sap/cds';

export default class AnalysisService extends cds.ApplicationService {
  async init() {
    const client = new OrchestrationClient({
      promptTemplating: {
        model: { name: 'gpt-4o' },
        prompt: [
          { role: 'system', content: 'Analyze and categorize as JSON.' },
          { role: 'user', content: '{{?input}}' }
        ]
      }
    });

    this.on('analyzeText', async (req) => {
      const response = await client.chatCompletion({
        placeholderValues: { input: req.data.text }
      });