> ## Documentation Index
> Fetch the complete documentation index at: https://autorender.io/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Replit Autorender Integration

> Add Autorender as the image infrastructure behind a Replit project, whether you're new to Autorender or already have a workspace.

[Replit](https://replit.com) is a platform for building sites and apps with AI. To add Autorender, paste one prompt into the Agent chat — the exact prompt depends on whether you already have an Autorender workspace.

## New to Autorender

<Steps>
  <Step title="Paste the onboarding prompt into Replit Agent's chat.">
    ```text theme={null}
    Read and follow https://autorender.io/agent-onboarding.md
    ```

    The agent reads that page and calls the Autorender API to provision an AI Agent Sandbox Account, then continues building against it. You don't create an account, configure a dashboard, or paste in an API key.

    <Info>
      Each sandbox includes **5 GB** of storage and **50 GB** of monthly bandwidth. Delivery is restricted to declared IP addresses.
    </Info>
  </Step>

  <Step title="Claim the workspace within 24 hours.">
    Once the agent finishes, it returns a claimable URL. Open it and sign in with Google or GitHub to keep the workspace.

    <Warning>
      Unclaimed sandboxes are deleted automatically **24 hours** after provisioning.
    </Warning>
  </Step>
</Steps>

## Already have an account

<Steps>
  <Step title="Paste this prompt into Replit Agent's chat.">
    ```text theme={null}
    Use Autorender (https://autorender.io) for image
    handling, optimization, and delivery.

    Ask me for my Autorender workspace ID first.

    If this app lets users upload images themselves —
    a file picker, drag-and-drop, or camera capture —
    ask me for my public key and use it directly in the
    browser or mobile code.

    For anything else — listing, renaming, deleting, or
    any server-side operation — ask me for a private key
    and store it as an environment variable on the
    server. Never expose a private key to the browser.

    Read https://autorender.io/docs/llms.txt for the
    integration reference.
    ```
  </Step>

  <Step title="Provide your workspace ID and API key when the agent asks.">
    Find both in your Autorender dashboard. A public key uploads only and is safe in browser code; a private key does everything else and must stay server-side. See [Create an API key](/docs/create-an-api-key) if you don't have one yet, or aren't sure which one your app needs.
  </Step>
</Steps>

## Next steps

<CardGroup cols={2}>
  <Card title="Create an API key" href="/docs/create-an-api-key">
    Manage public and private keys for your workspace.
  </Card>

  <Card title="MCP Server" href="/docs/mcp-server">
    Connect Autorender to Claude and other AI models over MCP.
  </Card>
</CardGroup>