Vercel AI SDK

1. Install the CLI

npm install -g swytchcode

irm https://cli.swytchcode.com/install.ps1 | iex

curl -fsSL https://cli.swytchcode.com/install.sh | sh

Use swy as a shorter alias for swytchcode.

2. Install the packages

JavaScript
Python

npm install @swytchcode/runtime ai @ai-sdk/openai dotenv

pip install swytchcode-runtime python-dotenv

3. Initialize Swytchcode and fetch a toolkit

swy init
swy get github

swy init creates .swytchcode/ and tooling.json for the project. swy get github downloads the GitHub integration bundle - the schemas that tools.get reads from in the next step.

import 'dotenv/config';
import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';
import { Swytchcode } from '@swytchcode/runtime';
import { VercelProvider } from '@swytchcode/runtime/providers/vercel';

const swx = new Swytchcode(new VercelProvider());
const tools = await swx.tools.get({ toolkits: ['github'] });

import os
from dotenv import load_dotenv
from swytchcode_runtime import Swytchcode

load_dotenv()

swx = Swytchcode()
tools = swx.tools.get(toolkits=["github"])

print(f"Loaded {len(tools)} tools")

5. Run the prompt

JavaScript
Python

const result = await generateText({
  model: openai('gpt-4.1'),
  tools,
  prompt: 'Summarize my open pull requests from GitHub.',
});

console.log(result.text);

result = swx.tools.execute("github.pull_requests.list", {
    "params": {"state": "open"},
    "Authorization": f"Bearer {os.getenv('GITHUB_TOKEN')}",
})

print(result)

6. Run the file

Save the example as main.js, then run:

node main.js

Swytchcode provides the tool execution layer while the SDK handles the model loop.