Skip to main content

API

The AGENTIC STAR API provides two families depending on your use case: the Agent API and the LLM Gateway.

What is the AGENTIC STAR API?

The Agent API enables real-time conversations with autonomous AI agents. The SaaS edition supports three connection methods (SSE, MCP, and A2A) and also provides REST APIs for user management. The Marketplace edition connects via Chat API / SSE. Conversation history and file management REST APIs are available in both editions.

The LLM Gateway is a raw LLM call (a governance proxy) compatible with OpenAI / Anthropic. You can use it by simply swapping the base URL of your existing OpenAI SDK / Anthropic SDK (SaaS only).

The Two API Families

FamilyUse CaseRepresentative Endpoint
Agent APIRunning autonomous agents (including tool execution and deliverable generation)/api/v1/chat/completions
LLM GatewayDirect OpenAI / Anthropic-compatible LLM calls/llm/v1/chat/completions
Which to Use

If you need agent features such as tool execution, deliverables, or conversation history, choose the Agent API. If you just want to call a raw LLM in an OpenAI / Anthropic-compatible way, choose the LLM Gateway.

Agent API

Getting Started

EditionSteps
SaaSCreate a client in Console → Try the API with the SaaS Quickstart
MarketplaceCreate a service account in Admin → Try the API with the Marketplace Quickstart

Guides / References

LLM Gateway

OpenAI / Anthropic-compatible LLM calls (SaaS only). Use it by swapping the base URL of your SDK.

  • LLM Gateway Quickstart — Shortest steps to call from an SDK
  • Setup — Preparation for administrators (model definition, IP) and developers (client creation)
  • LLM Gateway Reference — Specifications for all endpoints (OpenAI / Anthropic compatible)