> ## Documentation Index
> Fetch the complete documentation index at: https://upstash-docs-landing-rework.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

export const Catalog = ({product, type, featured, cols = 3}) => {
  const catalog = [{
    title: "@upstash/redis",
    description: "HTTP-based Redis client for serverless and edge.",
    href: "/redis/sdks/ts/overview",
    product: "redis",
    type: "sdk",
    lang: "typescript",
    featured: true
  }, {
    title: "upstash-redis (Python)",
    description: "Serverless Redis client for Python.",
    href: "/redis/sdks/py/overview",
    product: "redis",
    type: "sdk",
    lang: "python"
  }, {
    title: "@upstash/ratelimit",
    description: "Rate limiting for serverless, built on Redis.",
    href: "/redis/sdks/ratelimit-ts/overview",
    product: "redis",
    type: "sdk",
    lang: "typescript",
    featured: true
  }, {
    title: "ratelimit (Python)",
    description: "Rate limiting SDK for Python.",
    href: "/redis/sdks/ratelimit-py/overview",
    product: "redis",
    type: "sdk",
    lang: "python"
  }, {
    title: "@upstash/vector",
    description: "Vector database client for AI and LLM apps.",
    href: "/vector/sdks/ts/getting-started",
    product: "vector",
    type: "sdk",
    lang: "typescript",
    featured: true
  }, {
    title: "upstash-vector (Python)",
    description: "Vector database client for Python.",
    href: "/vector/sdks/py/gettingstarted",
    product: "vector",
    type: "sdk",
    lang: "python"
  }, {
    title: "@upstash/qstash",
    description: "Publish messages and schedule jobs over HTTP.",
    href: "/qstash/sdks/ts/overview",
    product: "qstash",
    type: "sdk",
    lang: "typescript",
    featured: true
  }, {
    title: "qstash-py",
    description: "QStash messaging and scheduling for Python.",
    href: "/qstash/sdks/py/overview",
    product: "qstash",
    type: "sdk",
    lang: "python"
  }, {
    title: "@upstash/workflow",
    description: "Durable serverless functions and workflows.",
    href: "/workflow/getstarted",
    product: "workflow",
    type: "sdk",
    lang: "typescript",
    featured: true
  }, {
    title: "@upstash/search",
    description: "Full-text and semantic search client.",
    href: "/search/sdks/ts/getting-started",
    product: "search",
    type: "sdk",
    lang: "typescript",
    featured: true
  }, {
    title: "BullMQ",
    description: "Message queue built on Redis.",
    href: "/redis/integrations/bullmq",
    product: "redis",
    type: "integration",
    lang: "typescript",
    featured: true
  }, {
    title: "Drizzle",
    description: "Cache Drizzle ORM queries with Upstash Redis.",
    href: "/redis/integrations/drizzle",
    product: "redis",
    type: "integration",
    lang: "typescript",
    featured: true
  }, {
    title: "Celery",
    description: "Use Upstash Redis as a Celery broker.",
    href: "/redis/integrations/celery",
    product: "redis",
    type: "integration",
    lang: "python"
  }, {
    title: "Sidekiq",
    description: "Background jobs for Ruby on Upstash Redis.",
    href: "/redis/integrations/sidekiq",
    product: "redis",
    type: "integration",
    lang: "ruby"
  }, {
    title: "MCP Server",
    description: "Manage Upstash Redis from AI agents via MCP.",
    href: "/redis/integrations/mcp",
    product: "redis",
    type: "integration",
    lang: "typescript"
  }, {
    title: "n8n (Redis)",
    description: "Automate workflows with Redis nodes in n8n.",
    href: "/redis/integrations/n8n",
    product: "redis",
    type: "integration",
    lang: "typescript"
  }, {
    title: "Prometheus (Redis)",
    description: "Scrape Upstash Redis metrics into Prometheus.",
    href: "/redis/integrations/prometheus",
    product: "redis",
    type: "integration",
    lang: "typescript"
  }, {
    title: "LangChain",
    description: "Vector store integration for LLM apps.",
    href: "/vector/integrations/langchain",
    product: "vector",
    type: "integration",
    lang: "python",
    featured: true
  }, {
    title: "Vercel AI SDK",
    description: "Back the AI SDK with Upstash Vector.",
    href: "/vector/integrations/ai-sdk",
    product: "vector",
    type: "integration",
    lang: "typescript",
    featured: true
  }, {
    title: "LlamaIndex",
    description: "Use Upstash Vector as a LlamaIndex store.",
    href: "/vector/integrations/llamaindex",
    product: "vector",
    type: "integration",
    lang: "python"
  }, {
    title: "Flowise",
    description: "Build LLM flows on Upstash Vector.",
    href: "/vector/integrations/flowise",
    product: "vector",
    type: "integration",
    lang: "typescript"
  }, {
    title: "Langflow",
    description: "Visual LLM apps backed by Upstash Vector.",
    href: "/vector/integrations/langflow",
    product: "vector",
    type: "integration",
    lang: "python"
  }, {
    title: "LlamaParse",
    description: "Parse and index documents into Upstash Vector.",
    href: "/vector/integrations/llamaparse",
    product: "vector",
    type: "integration",
    lang: "python"
  }, {
    title: "Resend",
    description: "Send emails from QStash workflows.",
    href: "/qstash/integrations/resend",
    product: "qstash",
    type: "integration",
    lang: "typescript",
    featured: true
  }, {
    title: "Anthropic",
    description: "Call Claude reliably through QStash.",
    href: "/qstash/integrations/anthropic",
    product: "qstash",
    type: "integration",
    lang: "typescript"
  }, {
    title: "LLM APIs",
    description: "Queue and retry LLM calls with QStash.",
    href: "/qstash/integrations/llm",
    product: "qstash",
    type: "integration",
    lang: "typescript"
  }, {
    title: "Datadog",
    description: "Monitor QStash with Datadog.",
    href: "/qstash/integrations/datadog",
    product: "qstash",
    type: "integration",
    lang: "typescript"
  }, {
    title: "Pipedream",
    description: "Trigger Pipedream workflows from QStash.",
    href: "/qstash/integrations/pipedream",
    product: "qstash",
    type: "integration",
    lang: "typescript"
  }, {
    title: "n8n (QStash)",
    description: "Publish QStash messages from n8n.",
    href: "/qstash/integrations/n8n",
    product: "qstash",
    type: "integration",
    lang: "typescript"
  }, {
    title: "Docusaurus",
    description: "Add full-text search to a Docusaurus site.",
    href: "/search/integrations/docusaurus",
    product: "search",
    type: "integration",
    lang: "typescript"
  }, {
    title: "AI SDK Code Interpreter",
    description: "Run AI-generated code in a Box sandbox.",
    href: "/box/guides/ai-sdk-code-interpreter",
    product: "box",
    type: "demo",
    lang: "typescript",
    featured: true
  }, {
    title: "E-commerce Order Fulfillment",
    description: "Durable order processing with Workflow.",
    href: "/workflow/examples/eCommerceOrderFulfillment",
    product: "workflow",
    type: "demo",
    lang: "typescript",
    featured: true
  }, {
    title: "Image Processing",
    description: "Fan-out image jobs with Workflow.",
    href: "/workflow/examples/imageProcessing",
    product: "workflow",
    type: "demo",
    lang: "typescript",
    featured: true
  }, {
    title: "Customer Onboarding",
    description: "Multi-step onboarding flow with delays.",
    href: "/workflow/examples/customerOnboarding",
    product: "workflow",
    type: "demo",
    lang: "typescript"
  }, {
    title: "Payment Retry",
    description: "Retry failed payments with backoff.",
    href: "/workflow/examples/paymentRetry",
    product: "workflow",
    type: "demo",
    lang: "typescript"
  }, {
    title: "Wait for Event",
    description: "Pause a workflow until an external event.",
    href: "/workflow/examples/waitForEvent",
    product: "workflow",
    type: "demo",
    lang: "typescript"
  }, {
    title: "Auth Webhook",
    description: "Handle auth webhooks durably.",
    href: "/workflow/examples/authWebhook",
    product: "workflow",
    type: "demo",
    lang: "typescript"
  }];
  const matchesProduct = entry => !product || (Array.isArray(entry.product) ? entry.product.includes(product) : entry.product === product);
  const items = catalog.filter(entry => matchesProduct(entry) && (!type || entry.type === type) && (!featured || entry.featured));
  const gridClass = "u-grid " + (cols >= 3 ? "u-grid--3" : "u-grid--2");
  return <div className={gridClass}>
      {items.map(entry => <a key={entry.href} href={entry.href} className="u-card">
          <img className="u-card__icon" src={`/img/icons/lang/${entry.lang}.svg`} alt={entry.lang} />
          <div className="u-card__body">
            <div className="u-card__title">{entry.title}</div>
            <div className="u-card__desc">{entry.description}</div>
          </div>
        </a>)}
    </div>;
};

**Upstash Box lets you give your AI agents a computer.**

Every Upstash Box is a **secure, isolated cloud container with an AI Agent built-in**. Spin up as many as you want in parallel. Each one includes a full environment with a filesystem, shell, git, and a runtime. Your agent can read files, write code, and execute tasks inside it.

<Note>Upstash Box is in developer preview — APIs and pricing may change.</Note>

### 1. Get your API key

Go to the [Upstash Console](https://console.upstash.com) and create an API key.

<Frame>
  <img src="https://mintcdn.com/upstash-docs-landing-rework/HSBmeqBTJDx6Qt9W/img/box/create-api-key.png?fit=max&auto=format&n=HSBmeqBTJDx6Qt9W&q=85&s=5e6fd736c59b7c100c168ea7771913bf" width="1131" height="844" data-path="img/box/create-api-key.png" />
</Frame>

### 2. Install the SDK

<CodeGroup>
  ```bash npm theme={null}
  npm install @upstash/box
  ```

  ```bash yarn theme={null}
  yarn add @upstash/box
  ```

  ```bash pnpm theme={null}
  pnpm add @upstash/box
  ```

  ```bash bun theme={null}
  bun install @upstash/box
  ```

  ```bash pip theme={null}
  pip install upstash-box
  ```
</CodeGroup>

### 3. Set API Key

```bash title=".env" theme={null}
UPSTASH_BOX_API_KEY=box_xxxxxxxxxxxxxxxxxxxxxxxx
```

### 4. Create a Box

<CodeGroup>
  ```typescript title="lib/box.ts" theme={null}
  import { Box } from "@upstash/box"

  const box = await Box.create({
    runtime: "node",
  })
  ```

  ```python box.py theme={null}
  from upstash_box import Box

  box = Box.create(runtime="node")
  ```
</CodeGroup>

<Tip>
  The Python SDK ships both a synchronous `Box` (used here) and an asynchronous
  `AsyncBox` (`box = await AsyncBox.create(...)`).
</Tip>

<Tip>
  By default, runtimes use **Debian** (glibc). For smaller Alpine-based images, use `"node-alpine"`, `"python-alpine"`, etc.
</Tip>

Your box is ready to use! You can already use it as a standalone, secure, isolated sandbox with full shell access, git, and filesystem operations.

<Tip>
  You can also create a keep-alive box by setting `keepAlive: true`. Keep-alive boxes stay on between sessions and can run an `initCommand` at startup. See [Keep Alive](/box/overall/keep-alive).
</Tip>

***

### 5. Configure an Agent (optional)

To configure an agent for your box, pass the model you want to use and the provider API key:

<Tabs>
  <Tab title="Claude">
    <CodeGroup>
      ```typescript box.ts {5-8} theme={null}
      import { Agent, Box } from "@upstash/box"

      const box = await Box.create({
        runtime: "node",
        agent: {
          harness: Agent.ClaudeCode,
          model: "anthropic/claude-opus-4-6",
          apiKey: process.env.ANTHROPIC_API_KEY,
        },
      })
      ```

      ```python box.py {6-10} theme={null}
      import os
      from upstash_box import Box, Agent

      box = Box.create(
          runtime="node",
          agent={
              "harness": Agent.CLAUDE_CODE,
              "model": "anthropic/claude-opus-4-6",
              "api_key": os.environ["ANTHROPIC_API_KEY"],
          },
      )
      ```
    </CodeGroup>
  </Tab>

  <Tab title="Codex">
    <CodeGroup>
      ```typescript box.ts {5-8} theme={null}
      import { Agent, Box } from "@upstash/box"

      const box = await Box.create({
        runtime: "node",
        agent: {
          harness: Agent.Codex,
          model: "openai/gpt-5.3-codex",
          apiKey: process.env.OPENAI_API_KEY,
        },
      })
      ```

      ```python box.py {6-10} theme={null}
      import os
      from upstash_box import Box, Agent

      box = Box.create(
          runtime="node",
          agent={
              "harness": Agent.CODEX,
              "model": "openai/gpt-5.3-codex",
              "api_key": os.environ["OPENAI_API_KEY"],
          },
      )
      ```
    </CodeGroup>
  </Tab>
</Tabs>

### 6. Run Your First Task

Your box is ready to use! It's a secure cloud environment to run agents, execute commands, or manage files. For example:

<CodeGroup>
  ```typescript box.ts theme={null}
  import { Agent, Box } from "@upstash/box"

  const box = await Box.create({
    runtime: "node",
    agent: {
      harness: Agent.ClaudeCode,
      model: "anthropic/claude-opus-4-6",
      apiKey: process.env.ANTHROPIC_API_KEY,
    },
  })

  // 👇 execute OS-level commands
  await box.exec.command("node --version")

  // 👇 run agent
  await box.agent.run({
    prompt: "create an index.txt saying 'hello world'",
  })
  ```

  ```python box.py theme={null}
  import os
  from upstash_box import Box, Agent

  box = Box.create(
      runtime="node",
      agent={
          "harness": Agent.CLAUDE_CODE,
          "model": "anthropic/claude-opus-4-6",
          "api_key": os.environ["ANTHROPIC_API_KEY"],
      },
  )

  # 👇 execute OS-level commands
  box.exec.command("node --version")

  # 👇 run agent
  box.agent.run(prompt="create an index.txt saying 'hello world'")
  ```
</CodeGroup>

### 7. Connect over SSH

You can also connect directly to a box shell with SSH:

```bash theme={null}
ssh <box-id>@us-east-1.box.upstash.com
```

When SSH asks for a password, enter your **Box API key**.

You can copy the exact SSH command for a box from the **SSH** button on its details page in the Upstash Console.

***

## Use Cases

The idea behind Upstash Box is simple: **give AI its own computer**. Your agent gets a full, isolated cloud environment it can control. Run commands, write files, or execute code independent of any user device. Freeze a box anytime, and continue days or even weeks later with perfect resumability.

Great example use cases:

<CardGroup cols={3}>
  <Card title="Agent Servers" icon="server" href="/box/overall/how-it-works#1-agent-server">
    One box per user with durable state. Personalized agents that remember context and improve over time.
  </Card>

  <Card title="Multi-Agent Orchestration" icon="diagram-project" href="/box/overall/how-it-works#2-multi-agent-orchestration">
    Fan out to multiple boxes running specialized agents in parallel, then combine their results.
  </Card>

  <Card title="Parallel Testing" icon="flask-vial" href="/box/overall/how-it-works#3-parallel-testing--model-comparison">
    Run the same inputs across isolated boxes and compare model output side by side.
  </Card>
</CardGroup>

***

## Next Steps

<CardGroup cols={2}>
  <Card title="How Boxes work" href="/box/overall/how-it-works">
    Learn the basics about using Upstash Box.
  </Card>

  <Card title="Agent" href="/box/overall/agent">
    Boxes have Claude Code, Codex or OpenCode built-in.
  </Card>
</CardGroup>

## Demos

<Catalog product="box" type="demo" cols={3} />

## Do more from the console

Watch your agents work, stream logs, and manage sandboxes from the dashboard.

<Tabs>
  <Tab title="CLI">
    <Note>
      Manage and connect to your boxes from the terminal. Screenshot coming soon.
    </Note>
  </Tab>

  <Tab title="Logs">
    <Note>
      Stream logs from your running boxes in the console. Screenshot coming soon.
    </Note>
  </Tab>
</Tabs>