Skip to main content

GitHub Repository

You can find the project source code on GitHub.
This guide provides detailed, step-by-step instructions on how to use Upstash Workflow with TanStack Start. You can also explore the source code for a detailed, end-to-end example and best practices.

Prerequisites

  • Node.js and pnpm installed.
You can integrate Upstash Workflow into an existing TanStack Start app, or create a new TanStack Start project from scratch.

Step 1: Create a new TanStack Start project

Run the following command to create a new TanStack Start project:
pnpm create @tanstack/start@latest
Navigate to the project directory:
cd your-project-name

Step 2: Installation

Run the following command to install the Upstash Workflow SDK in your TanStack Start app.
  • pnpm
  • npm
  • bun
pnpm install @upstash/workflow

Step 3: Run the development server

Upstash Workflow is built on top of Upstash QStash. In a production environment, your application connects to the managed QStash servers hosted by Upstash. This ensures that requests are delivered reliably, securely, and at scale without requiring you to run and maintain your own infrastructure. For local development, you don’t need to depend on the managed QStash servers. Instead, you can run a local QStash server directly on your machine. This local server behaves just like the production version but does not require external network calls. Start the local server with:
  • pnpm
  • npm
pnpx @upstash/qstash-cli dev
When the server starts, it will print the credentials. You’ll need these values in the next step to connect your TanStack Start app to QStash. You can enable local mode in the Upstash Workflow dashboard to use the UI while developing locally.
Enable local mode on dashboard

Step 4: Configure Environment Variables

Next, you need to configure your TanStack Start app to connect with the local QStash server by setting environment variables. In the root of your project, create a .env file (or update your existing one) and add the values printed by the QStash local server:
QSTASH_URL="http://127.0.0.1:8080"
QSTASH_TOKEN="eyJVc2VySUQiOiJkZWZhdWx0VXNlciIsIlBhc3N3b3JkIjoiZGVmYXVsdFBhc3N3b3JkIn0="
QSTASH_CURRENT_SIGNING_KEY="sig_7kYjw48mhY7kAjqNGcy6cr29RJ6r"
QSTASH_NEXT_SIGNING_KEY="sig_5ZB6DVzB1wjE8S6rZ7eenA8Pdnhs"
For production, replace these with your actual credentials from the Upstash Workflow dashboard.

Step 5: Create a Workflow Endpoint

With your environment ready, the next step is to define your first workflow endpoint. In Upstash Workflow, every workflow is exposed as an endpoint. Every endpoint you expose using the SDK’s serve() function acts as a workflow that can be triggered independently. In TanStack Start, these endpoints are implemented as API routes using the file-based routing system. Create a new file src/routes/api/workflow.ts:
src/routes/api/workflow.ts
import { createFileRoute } from '@tanstack/react-router'
import { serve } from '@upstash/workflow/tanstack'

const someWork = (input: string) => {
  return `processed '${JSON.stringify(input)}'`
}

export const Route = createFileRoute('/api/workflow')({
  server: {
    handlers: serve<string>(async (context) => {
      const input = context.requestPayload
      
      const result1 = await context.run('step1', () => {
        const output = someWork(input)
        console.log('step 1 input', input, 'output', output)
        return output
      })

      await context.run('step2', () => {
        const output = someWork(result1)
        console.log('step 2 input', result1, 'output', output)
      })
    }),
  },
})

Step 6: Start your TanStack Start app

Start your TanStack Start development server:
  • pnpm
  • npm
pnpm dev
Your app should now be running on http://localhost:3000.

Step 7: Run the Workflow Endpoint

Once your endpoint is defined, the next step is to trigger a workflow run. You can start a new workflow run using the trigger() function from the Upstash Workflow SDK.
import { Client } from "@upstash/workflow";

const client = Client()

const { workflowRunId } = await client.trigger({
    url: `http://localhost:3000/api/workflow`,
    body: "Hello World!",
    retries: 3,
});
The trigger() function should typically be called from a server-side action (not directly in client-side code) to keep your credentials secure.
Check the Upstash Workflow dashboard to view logs of your workflow run:
Debug a workflow run on UI
Inside the trigger() call, you need to provide the URL of your workflow endpoint:To avoid hardcoding URLs, you can define a BASE_URL constant and set it based on the environment:
const BASE_URL = process.env.NODE_ENV === 'production'
  ? 'https://yourapp.com'
  : 'http://localhost:3000'

const { workflowRunId } = await client.trigger({
    url: `${BASE_URL}/api/workflow`,
    body: "Hello World!",
    retries: 3,
});

Next Steps

Now that you’ve created your first workflow, here are some recommended guides to continue learning:
  1. Learn the Workflow API: Dive deeper into the full API surface and advanced capabilities.
  2. Configure Workflow Runs: Learn how to configure workflow execution to fit your app’s needs.
  3. Handle Failures: Understand how to detect and recover from failed workflow runs.
I