CapabilityAI & AgentsNode.js, read from its artifacts

Prompt management

Keeping 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.

Showing artifacts, examples and content with confirmed Node.js applicability. View All CAP

Overview

SAP SHIPS

No SAP artifact is confirmed here for Node.js.

A Node.js coverage assessment has not been written. View all flavors

THE COMMUNITY ADDS
@saptarishi/cds-plugin-llm

A prompt registry that lives in the application rather than in a service: templates are registered once and invoked by name from a CAP handler, gitPromptRegistry loads them from a Git repository and caches them instead, so a wording change goes through pull request review separately from a code deployment, and promptVersionPin decides which version a call or a session gets.

A Node.js coverage assessment has not been written. View all flavors

Artifacts2

2 of 2 artifacts
ArtifactSignals
CDS plugins1
@saptarishi/cds-plugin-llmCAP service kind putting eleven model providers behind one API
Signals
none
Services1
SAP AI CoreAI runtime on BTP with the Generative AI Hub
Signals
none

Seen working2

ArtifactSignals
Samples2
SAP BTP AI Best Practice Demo - Prompt Registry - CAPCAP samples using SAP AI Core's Prompt Registry for templates
Functional boundary
SAP BTP AI Best Practice Demo - Prompt Templating - CAPCAP samples filling a prompt template sent to SAP AI Core
Functional boundary

Details

Coverage
statecommunity only, CAP, Node.js, derived
also inBTP/BAIP
For your agent
Ref
capability:prompt-management
JSON
/capabilities/prompt-management.json
MCP
Connecting an agent