Agentuity v1 is Now in Public Preview

Today, we're excited to open up the Agentuity v1 Platform for public preview. This milestone represents countless hours working with customers and developers on what the best end-to-end agentic developer experience should be. We've learned a lot since we launched our early access version back in the summer and we're excited to get this in everyone's hands.

v1 represents a new foundation for building agents with full-stack features:

• End-to-end type safety — from agent typing and API typing down to React hooks
• Ability to define APIs separate from agents — with helpers like router.stream(), router.email(), router.cron(), and more
• Flexible programming model — build things the way you're used to
• Ability to serve frontends — deploy directly alongside your agents or host separately
• Frontend SDK — such as React hooks like useAPI() and useWebsocket() that make it a breeze to work across the stack
• First-class Bun support — JavaScript is the language of agents and Bun is the future of JavaScript, so we're all in on Bun
• Built-in services — S3-compatible storage, managed Postgres, KV, durable streams, and more
• And more to come...

This is an early alpha release, so give it a spin and tell us what works (and what doesn't). The SDK is also open source, so we'd love your contributions. We're looking for some collaboration specifically around frameworks like Svelte.

Try It Out

Watch the walkthrough to see the full setup flow:

Or jump straight in — here are the steps:

1. Install the CLI:

curl -sSL https://v1.agentuity.sh | sh

2. Once that's installed, run agentuity create to scaffold a new project.

agentuity create

3. cd into the project and run agentuity dev to get started.

cd my-project
agentuity dev

For examples, SDK reference, and migration guide, check out the v1 docs.

A Quick Example: Full-Stack in Action

Here's what a complete agent looks like — from backend logic to API route to React frontend:

1. The Agent (src/agent/chat/agent.ts)

import { createAgent } from '@agentuity/runtime';
import { s } from '@agentuity/schema';
import { generateText } from 'ai';
import { openai } from '@ai-sdk/openai';

const agent = createAgent('chat', {
  description: 'A simple chat agent',
  schema: {
    input: s.object({ message: s.string() }),
    output: s.object({ response: s.string() }),
  },
  handler: async (ctx, { message }) => {
    const { text } = await generateText({
      model: openai('gpt-5-mini'),
      prompt: message,
    });
    return { response: text };
  },
});

export default agent;

2. The API Route (src/api/index.ts)

import { createRouter } from '@agentuity/runtime';
import chatAgent from '@agent/chat';

const router = createRouter();

router.post('/chat', chatAgent.validator(), async (c) => {
  const data = c.req.valid('json');
  const result = await chatAgent.run(data);
  return c.json(result);
});

export default router;

3. The Frontend (src/web/components/Chat.tsx)

import { useAPI } from '@agentuity/react';

function Chat() {
  const { invoke, data, isLoading } = useAPI('POST /api/chat');

  return (
    <div>
      <button onClick={() => invoke({ message: 'Hello!' })}>
        {isLoading ? 'Thinking...' : 'Send'}
      </button>
      {data && <p>{data.response}</p>}
    </div>
  );
}

Feedback, Support, and What's Next

This is a public preview, and we're targeting a full release in early January. Your feedback during this preview will directly shape what ships.

Here's how to get involved on GitHub:

• Found a bug? Open an issue
• Have a question? Post in Q&A
• Have an idea? Share it in Ideas
• Built something? Show us in Show and tell

Resources

• Check out the Agentuity v1 docs for more examples, SDK reference, and migration guide
• Access the Web Console to manage your v1 projects
• Join us on Discord to chat with the team and community

We can't wait to see what you build and hear what you think!