Run openclaw securely without any dedicated mac-mini, hostinger or gpu.
Brew your lobster securely without breaking your bank
OpenClaw-WebTop gives you a fully functional OpenClaw personal AI assistant in your browser in under 20 minutes — no powerful PC, no Docker on your machine, no GPU required.
Just open this repo in a GitHub Codespace and you get:
- A complete Ubuntu MATE desktop (WebTop)
- Ollama server pre-installed and auto-started
- OpenClaw globally installed via npm
- Persistent volume for your config, pairings, and OpenClaw ID
When you’re ready to go production, simply move the same Docker setup to your own machine or VPS. *
OpenClaw (the core project) is one of the most exciting AI agent frameworks right now — it connects LLMs directly to your WhatsApp, Telegram, Slack, Discord, etc., and can run cron jobs, spawn sub-agents, speak/listen, and give you a beautiful dashboard.
The only catch? You normally need a dedicated machine.
OpenClaw-WebTop removes that catch completely.
Perfect for:
- Trying OpenClaw risk-free
- Students / hackers / evaluators
- Anyone who wants to “brew their lobster” securely on free cloud credits
Demo of the full process
Secured.OpenClaw.WebTop-small.mp4
-
Open this repository in a GitHub Codespace (big green “Code” button → Codespaces → New)

-
In the Codespace terminal run:
make start
(or make start-locally-baked if you prefer a pre-built image)
-
Wait ~60 seconds. When the web desktop URL appears in the Codespace Ports tab, click it.

-
Inside the WebTop desktop:
- Open a terminal →
ollama signin(sign in via the Chromium browser that pops up)
- Pull a model:
ollama pull kimi-k2.5:cloud(or any model you like) - Launch:
ollama launch openclaw --model kimi-k2.5:cloud --yes(Use this command for fresh codespace) - (If there is errors/After first launch)
openclaw gateway runoropenclaw gateway restart - Finally:
openclaw dashboard→ copy the tokenized URL
- Open Chromium inside WebTop and paste the dashboard URL.
You now have a fully working OpenClaw instance running 100% in the cloud.

- Zero local install — everything runs in browser via GitHub Codespaces
- Free-tier friendly — uses Ollama daily cloud credits + NVIDIA Build API fallback
- Persistent config — if docker volume backup and restore after Codespace recreation
- Easy backup/restore —
make backup/make restore - One-command everything — powerful Makefile + clean
docker-compose.yml - Auto-start Ollama — custom init script on WebTop boot
- NVIDIA Build fallback built-in
- Colima / local Docker support ready
The WebTop URI is automatically protected — no one else can reach it.
GitHub Codespaces forwards ports privately by default (this is the setting the make start command uses). According to official GitHub documentation:
“All forwarded ports are private by default, which means that you will need to authenticate before you can access the port.”
“Privately forwarded ports: Are accessible on the internet, but only the codespace creator can access them, after authenticating to GitHub.”
- The URL you click in the Ports tab (
https://<your-codespace>-3000.app.github.dev) is guarded by GitHub authentication cookies. - These cookies expire every 3 hours — you’ll simply be asked to log in again (super quick).
- If someone tries to open the link in an incognito window, via curl, or from another computer without being logged into your GitHub account, they are redirected to the GitHub login page or blocked.
- You (and only you) can access the full Ubuntu desktop, the browser inside it, Ollama, OpenClaw dashboard, and everything else.
- The entire environment runs in an isolated GitHub-managed VM — not on your laptop.
- Codespaces are ephemeral: delete the codespace and everything disappears (except the backed-up volume you control).
- TLS encryption is handled automatically by GitHub.
- The
GITHUB_TOKENinside the codespace is scoped only to this repo and expires when you stop/restart. - We never set the port to “Public” or even “Private to Organization” — it stays strictly private to you.
Bottom line: This is actually more secure for experimentation than running Docker locally on your personal machine (no accidental exposure, no firewall holes, no persistent processes on your hardware).
For production use we still recommend moving the same Docker image to your own VPS or server with additional hardening (firewall, HTTPS reverse proxy, strong secrets, etc.). This Codespace version is perfect for safe testing and development.
Your OpenClaw ID, device pairings, and configuration are persisted in a Docker volume.
The project includes convenient make targets to back up and restore this data in codespace:
make backup # creates backup/openclaw-webtop.tar.gz
make restore # restores from backup/openclaw-webtop.tar.gz- Migrating from GitHub Codespaces to a local machine or VPS
- Testing experimental changes without risking your current setup
- Quickly cloning your working environment into a fresh Codespace or container
- In your current environment, run
make backup. - Download the generated file:
backup/openclaw-webtop.tar.gz. - Place the file in the
backup/folder of the new environment. - Run
make restore.
💡 Tip: Always back up before making significant changes. The restore process will overwrite the existing volume data, so test in a separate environment first if you're unsure.
Run locally (no Codespaces)
make build-local # especially if you modified the ./docker/Dockerfile
make start-locally-baked # start from your local bake imageNVIDIA offers free inference endpoints for powerful models like Moonshot AI's Kimi K2 (e.g., kimi-k2-instruct) through their Build platform. These make an excellent low-cost or zero-cost fallback when your primary model hits rate limits or goes down.
- Go to NVIDIA Build and sign in (free NVIDIA Developer account required).
- Navigate to the Models tab and find a Free Endpoint. Good choices:
moonshotai/kimi-k2-instruct - Click on the model. This opens a chat interface.
- Click the View Code button (usually top-right).
- In the "Copy Code to Make an API Request" panel:
- Click Generate API Key — the key will automatically appear in the code sample.
- Copy the entire Python code snippet.
- Paste the code into a chat with OpenClaw and use this prompt:
Hey, here's the Python code from NVIDIA Build for the Kimi model (with my API key already included).
1. Test the code to make sure it works.
2. Configure this model as a fallback for my primary model.
- OpenClaw should automatically detect the OpenAI-compatible endpoint, extract the key, and set it up as a fallback.
- GitHub Codespaces free tier has monthly limits (great for testing, less ideal for 24/7 as Codespace auto-shutdown during inactivity)
- Ollama cloud credits are daily — heavy use will push you to paid/local models. Or if you have multiple accounts, just
ollama signoutandollamasigninwith different account. - Browser desktop has slight latency vs native (expected). You can shutdown your codespace and change to 4-core codespace to improve responsiveness or the need to run heavy applications.
- More screenshots + video demo
- Full pairing automation scripts
- Pre-built Docker image tags for stable releases
- Community templates (Telegram-only, WhatsApp-only, etc.)
- One-click “deploy to VPS” guide (Railway / Fly.io / cheap VPS) ?
This is a one-person weekend project right now — every star, issue, or PR helps enormously! Feel free to open issues for bugs or feature requests.
Apache-2.0 — see LICENSE
