OpenAI Function Calling

Use y0 with OpenAI GPT models via function calling.

Overview

While y0’s primary protocol is MCP, you can also use the tools via OpenAI’s function calling format. A JS SDK with pre-built adapters is planned.

Using with OpenAI SDK (Coming Soon)

import OpenAI from 'openai'
import { y0Tools } from '@y0exchange/sdk-js/openai'
 
const openai = new OpenAI()
 
const response = await openai.chat.completions.create({
  model: 'gpt-4',
  messages: [
    { role: 'user', content: 'What is the ETH balance of 0xd8dA...?' }
  ],
  tools: y0Tools,
})

Tool Definitions

The y0 tools map directly to OpenAI function calling format:

{
  "type": "function",
  "function": {
    "name": "get_balance",
    "description": "Get native and ERC-20 token balances for a wallet address",
    "parameters": {
      "type": "object",
      "properties": {
        "address": {
          "type": "string",
          "description": "Wallet address (0x...)"
        },
        "chainId": {
          "type": "number",
          "description": "Chain ID (default: 1)"
        }
      },
      "required": ["address"]
    }
  }
}

Notes

A JS SDK (@y0exchange/sdk-js) with pre-built OpenAI tool definitions is planned
Currently, the MCP server is the recommended integration method
Write tools work the same way — they build unsigned transactions for user signing
The signing flow is identical regardless of the AI framework used

LangChain / Python Authentication