API Reference Overview

The Vertical Template provides multiple API surfaces for building multi-tenant, config-driven vertical applications on the EnterpriseAI platform. This reference covers the Platform SDK, React Hooks, REST API, configuration schemas, and UI components.

Quick Reference

API Surface	Purpose	Best For
Platform SDK	Type-safe JavaScript/TypeScript client	Backend services, Node.js apps
React Hooks	Client-side data fetching and state	Frontend components, React apps
REST API	Direct HTTP access	cURL, external integrations
Config Schema	Tenant configuration structure	JSON config files, admin tools
Components	Pre-built UI components	React applications
Interactive Tools	Dynamic UI widgets	Chat, document upload, actions

Platform SDK

The @enterpriseaigroup/platform-sdk package provides a comprehensive TypeScript SDK for interacting with EnterpriseAI services.

Installation

npm install @enterpriseaigroup/platform-sdk
# or
yarn add @enterpriseaigroup/platform-sdk

Quick Start

import { PlatformClient } from '@enterpriseaigroup/platform-sdk';

const client = new PlatformClient({
  apiKey: process.env.EAI_API_KEY,
  baseURL: 'https://api.enterpriseai.com',
  tenantId: 'your-tenant-id'
});

// Use SDK modules
const resources = await client.resources.list({ type: 'project' });
const chatResponse = await client.chat.sendMessage('Hello!');

SDK Modules

• Client - Initialization and configuration
• Resources - CRUD operations on typed resources
• Chat - SSE streaming chat with AI
• Documents - File upload, classification, RAG indexing
• Users - User profile management
• Auth - Token management and authentication
• Orchestrate - Workflow execution

React Hooks

The @enterpriseaigroup/core package provides React hooks for seamless integration with EnterpriseAI services in your frontend.

Installation

npm install @enterpriseaigroup/core
# or
yarn add @enterpriseaigroup/core

Quick Start

import { useResources, useChat } from '@enterpriseaigroup/core';

function MyComponent() {
  const { data: projects, loading } = useResources({ type: 'project' });
  const { messages, sendMessage } = useChat();

  return (
    <div>
      {projects?.map(p => <div key={p.id}>{p.name}</div>)}
    </div>
  );
}

Available Hooks

• useResources - Typed CRUD for resources with caching
• useChat - SSE streaming chat with real-time updates
• useDocuments - File upload and document management

REST API

All API calls in the Vertical Template use a BFF (Backend-for-Frontend) proxy pattern. Client requests go through /api/eai/* routes where authentication tokens are injected server-side.

Base URL

https://your-app.vercel.app/api/eai/v3

Authentication

All requests require an Authorization header with a Bearer token. In the Vertical Template, this is handled automatically by the BFF proxy.

curl -H "Authorization: Bearer YOUR_TOKEN" \
  https://your-app.vercel.app/api/eai/v3/resources

API Endpoints

• Resources API - GET, POST, PUT, DELETE resources
• Chat API - POST streaming chat (SSE)
• Documents API - POST multipart uploads, GET, DELETE
• Users API - GET user profiles
• Orchestrate API - POST workflow execution

Configuration

The Vertical Template is config-driven, allowing you to define layouts, components, and data bindings in JSON configuration files.

Configuration Files

• Config Schema - Complete EAIConfig TypeScript type reference
• Components - Built-in component registry and custom registration
• Interactive Tools - ActionBar, ChatPanel, DocumentUploader

Example Configuration

{
  "layout": {
    "header": {
      "component": "Header",
      "props": {
        "logo": "/logo.png",
        "title": "My Vertical App"
      }
    },
    "slots": [
      {
        "id": "main",
        "component": "ResourceList",
        "store": {
          "binding": "resources.projects"
        }
      }
    ]
  }
}

Backend Services

The Vertical Template integrates with the following EnterpriseAI backend services:

Service	Purpose	SDK Module
PublicAPI	Resource CRUD, chat, documents	resources, chat, documents
Configurator	Tenant configuration management	Loaded via getConfig()
ResourceAPI	Advanced resource queries	resources.query()
AICore	LLM orchestration and RAG	chat, orchestrate
Authz	Authorization and access control	auth, headers

TypeScript Support

All SDK modules and hooks are fully typed with TypeScript. Use generic type parameters for typed responses:

interface Project {
  id: string;
  name: string;
  status: 'active' | 'archived';
}

// SDK
const projects = await client.resources.list<Project>({ type: 'project' });

// Hooks
const { data } = useResources<Project>({ type: 'project' });

Error Handling

All SDK methods and hooks return structured errors:

try {
  const resource = await client.resources.create(data);
} catch (error) {
  if (error.code === 'VALIDATION_ERROR') {
    console.error('Invalid data:', error.details);
  } else if (error.code === 'UNAUTHORIZED') {
    console.error('Authentication failed');
  }
}

Rate Limits

The EnterpriseAI platform enforces the following rate limits:

• Resources API: 1000 requests/minute per tenant
• Chat API: 100 concurrent streams per tenant
• Documents API: 50 uploads/minute per tenant

Rate limit headers are included in all responses:

X-RateLimit-Limit: 1000
X-RateLimit-Remaining: 847
X-RateLimit-Reset: 1678886400

Next Steps

• Getting Started: See the Quick Start Guide
• Guides: Explore integration patterns
• Examples: Check out the example applications
• Support: Join our Discord community