Overview

Make your product AI-powered without prompts. Functions let you call curated LLM skills from code and receive text or JSON outputs you can trust, with versioning, tests, and observability built in.

📘

Use Functions when you want a deterministic API (well-defined inputs/outputs) instead of an open-ended chat.

When to Use Functions

• Synchronous AI APIs: quick, reliable responses for product features.
• Deterministic contracts: predictable outputs via schemas.
• Non-chat use cases: summaries, classification, extraction, or generation tasks.

Prerequisites

📘

⚠️ Important: Functions are created and versioned in the Agent Studio UI under Actions → Functions.

How it works

1. In Agent Studio (UI)

   • Define a new Function under Actions → Functions.
   • Provide input/output schema, description, and optional test cases.
   • Save and Publish a version (e.g. v1).

2. In your app (code)

   • Call the Function by name with its published schema.
   • Handle success/failure and use the structured result.

Parameters

Using aiFunctionCall

const apiKey: string = "generate_xyz";
// Input
const data: string | number | object | Array<any> = {};

// Callback response
const resolve = (value: object) => {
  console.log("Success", value);
};

const reject = (reason?: any) => {
  console.error("Error", reason);
};

// Call the API
window.eucera.agent("your-agent-api-name").aiFunctionCall(apiKey, data, resolve, reject);