OpenAI Integration

Note: The llmConfig parameters correspond to the underlying LLM API you’re using. For detailed parameter information, check:

OpenAI: Chat Completions API Reference

Important: Make sure to handle streaming responses appropriately when implementing tool usage.


import OpenAI from "openai";
import { OpenAIAdapter } from "@reacter/openapitools";
 
// Initialize the AI client
const openai = new OpenAI({
  apiKey: "your-openai-api-key",
});
 
// Initialize the adapter
const adapter = new OpenAIAdapter(
  "your OpenAPI Tools Apikey - https://openapitools.com/dashboard/settings",
  {
    verbose: true,
  }
);
 
// Initialize the adapter first
await adapter.initialize();
 
// Create a chatbot with tools
const chatbot = await adapter.createOpenAIChatbot({
  openaiClient: openai,
  llmConfig: {
    model: "gpt-4o",
    temperature: 0.7,
    max_tokens: 4096,
    system: "You are a helpful assistant with access to tools.",
  },
  options: {
    // Optional
    toolNames: ["generate-otp", { name: "get-orders", version: "prod" }],
  },
});
 
// Use the chatbot
const result = await chatbot.invoke("Can you help me check the weather?");
console.log(result.text);