Use OpenAI SDK models with Swytchcode tools

Build OpenAI SDK agent flows that call Swytchcode toolchains.

Book a demo

Quickstart steps

Standard SDK path

Tool-safe

Execution

Fast

Developer onboarding

CLI

Ship in one command.

Pull the manifest, set your policy, execute. No SDK sprawl, no boilerplate. Just the tool you need, right now.

npm install @swytchcode/runtime openai dotenv
swytchcode login
swytchcode get github

How it connects

Connect OpenAI SDK in minutes

Follow these steps to connect your agent, load a toolkit, and start executing tools through Swytchcode.

1
Install the packages
Install Swytchcode runtime and the OpenAI SDK packages.
```
npm install @swytchcode/runtime openai dotenv
```
2
Sign in
Authenticate with Swytchcode so the runtime can fetch tools.
```
swytchcode login
```

Create agent tools

Initialize Swytchcode with the OpenAI provider and load the toolkit.

import { Swytchcode } from '@swytchcode/runtime'
import { OpenAIProvider } from '@swytchcode/runtime/providers/openai'
const swx = new Swytchcode(new OpenAIProvider())
const tools = await swx.tools.get({ toolkits: ['github'] })

Run the model

Call the OpenAI SDK with `tools` so Swytchcode handles runtime execution.

const response = await client.responses.create({
  model: 'gpt-4.1',
  tools,
  input: 'List my open GitHub issues and summarize them.',
})

5
Execute the file
Run your app and let Swytchcode turn the model's tool calls into safe API actions.
```
node main.js
```

Works with

APIs your team already uses

GitHub

Stripe

Ready to ship openai sdk integrations?

Install the CLI and pull your first manifest in under a minute. No account required.