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

# OpenClaw Template on Vertra Cloud: Personal AI Assistant with a Browser

> What the OpenClaw template creates on Vertra Cloud: memory, what it asks, channels, variables and license. The full walkthrough is in the guide.

A personal AI assistant with a control panel, messaging channels and an optional browser. The template runs OpenClaw 2026.10.7 with headless Chromium inside the container. You bring the model key: Vertra Cloud does not host a model.

> [card] [Step-by-step guide: OpenClaw from the key to your channels](https://docs.vertracloud.app/templates/guides/openclaw.md)
>
> Provider and model, the gateway token, approving devices, Telegram, Discord, Slack and WhatsApp, and the browser.

## What it creates

One app (image) with the panel at the app address. **Memory: 3 GB minimum with the browser** (4 GB recommended) or 2 GB without it (3 GB recommended). It needs a plan with Docker images and one free [image slot](https://docs.vertracloud.app/templates/overview.md#image-slots); see [Plans and limits](https://docs.vertracloud.app/plans-and-limits.md).

## What the wizard asks

| Question | What it does |
| - | - |
| Model provider | OpenAI, Anthropic, Google Gemini, OpenRouter or an OpenAI-compatible endpoint. |
| Provider API key | Kept only in the app's encrypted variables. |
| Endpoint base URL | Only with a compatible endpoint; ends in `/v1`. |
| Model (optional) | Empty means the provider's default; required with a compatible endpoint. |
| Can the assistant use a browser? | Yes (default) or no. Applies on the first start. |
| Telegram, Discord, Slack | Token and who may talk to the bot. Without an allowed list the channel is not enabled. |

## What you get

The app address, the **gateway token** (shown once) and the next step. WhatsApp is paired by QR code in the panel.

## Variables

| Variable | What it does |
| - | - |
| `OPENCLAW_GATEWAY_TOKEN` | The credential for the panel. |
| `PUBLIC_ORIGIN` | The app's public address. |
| `OPENCLAW_PROVIDER`, `OPENCLAW_MODEL`, `OPENCLAW_MODEL_ID` | Provider and model, recomputed on every start. |
| `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, `GEMINI_API_KEY`, `OPENROUTER_API_KEY`, `OPENAI_COMPAT_API_KEY` | The chosen provider's key; `OPENAI_COMPAT_BASE_URL` holds the compatible endpoint. |
| `OPENCLAW_BROWSER` | `on` or `off`. |
| `OPENCLAW_BOOTSTRAP_PAIRING` | `off` turns off the automatic approval of the first browser. |

## Resources and limits

* Port 18789. First boot 30 to 45 seconds, and 1 to 3 minutes with Discord or Slack.
* State in `/app/.openclaw`; channel plugins from npm add about 240 MB.
* Zero-downtime deploy is off; rollback returns the data too.
* The agent browser blocks private and local networks.

## License

OpenClaw is distributed under the [MIT license](https://github.com/openclaw/openclaw/blob/main/LICENSE). The Chromium in the image follows the Chromium project's license. This template is a Vertra Cloud packaging and is not affiliated with the project.

See also the [public page](https://vertracloud.app/templates/openclaw) and the [other official templates](https://docs.vertracloud.app/templates/catalog.md).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.