Overview
OpenClaw AI Workstation for OpenClaw 2.0 is a ready-to-run Ubuntu 24.04 AMI for teams and builders who want a private cloud AI workstation that can run 24/7 in their own AWS account.
This image is a fresh OpenClaw 2.0 installation, not an in-place upgrade image. It combines OpenClaw 2026.8.1, Hermes Agent, Open WebUI, Ollama, browser automation, secure networking, and developer tooling into one opinionated virtual machine. Launch the AMI, SSH in, run agent-menu, and choose whether to use OpenClaw or Hermes as the active agent.
What is included:
- OpenClaw 2026.8.1
- Hermes Agent v0.20.0 based on upstream v2026.8.3
- agent-menu for switching between OpenClaw and Hermes
- Open WebUI for local/private chat and Ollama model access
- Ollama local model runtime
- Google Chrome, Playwright, Xvfb, x11vnc, and noVNC for browser automation
- Docker Engine with Docker Compose and Buildx
- Caddy web server for HTTPS reverse proxy setups
- Tailscale for private networking and secure remote access
- Developer tools including GitHub CLI, AWS CLI, jq, yq, ripgrep, fd-find, ffmpeg, tmux, and htop
The AMI is bring-your-own-model-access. No AI provider credentials, API keys, tokens, models, or subscriptions are included. Users can configure OpenRouter, OpenAI, Anthropic, local Ollama models, or other supported OpenAI-compatible APIs.
Security model:
- SSH is the primary management interface.
- noVNC is available for visual browser automation; restrict access with security groups, SSH tunnels, Caddy, or Tailscale.
- Open WebUI and Ollama bind locally by default for safer access via SSH tunnels or private networking.
- Root SSH login and SSH password authentication are disabled.
This product packages open source software for convenience and is not affiliated with or endorsed by OpenClaw, Hermes Agent, Nous Research, Open WebUI, Ollama, Tailscale, Caddy, Docker, OpenAI, Anthropic, GitHub, or AWS.
Highlights
- Fresh OpenClaw 2.0 workstation image with OpenClaw 2026.8.1, Hermes Agent, Open WebUI, Ollama, Docker, Caddy, and Tailscale preinstalled
- agent-menu lets users choose OpenClaw or Hermes as the active agent after first SSH login
- Private AI workstation design: local Ollama support, local-only app ports, and secure access via SSH tunnels, HTTPS, or Tailscale
Details
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
Pricing
- ...
Dimension | Cost/hour |
|---|---|
t3a.large Recommended | $0.04 |
t3.micro | $0.03 |
t2.micro | $0.03 |
c6i.24xlarge | $0.05 |
m8a.2xlarge | $0.05 |
m5a.16xlarge | $0.05 |
m6a.48xlarge | $0.05 |
c7i.12xlarge | $0.05 |
c8i.xlarge | $0.05 |
c8i.4xlarge | $0.05 |
Vendor refund policy
Usage-based hourly billing. No commitment, cancel anytime by stopping your EC2 instance. You only pay for hours consumed. If the software fails to function as described, contact support@breakingit.nl within 48 hours for a refund of the affected period.
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
64-bit (x86) Amazon Machine Image (AMI)
Amazon Machine Image (AMI)
An AMI is a virtual image that provides the information required to launch an instance. Amazon EC2 (Elastic Compute Cloud) instances are virtual servers on which you can run your applications and workloads, offering varying combinations of CPU, memory, storage, and networking resources. You can launch as many instances from as many different AMIs as you need.
Version release notes
OpenClaw updated to exactly 2026.9.6; Ubuntu 24.04 packages refreshed. Retains the previously published AI Workstation tools and security configuration, including Hermes Agent, Open WebUI, Ollama, sandboxed Chrome/Playwright, noVNC, Docker, Tailscale and Caddy. Application services bind to loopback; use SSH tunnels. No AI provider credentials or preloaded models included.
Additional details
Usage instructions
Launch an instance (t3a.large or larger recommended for the full workstation; local models need additional RAM). Restrict SSH to your administrator IP.
Connect and tunnel local services: ssh -L 18789:127.0.0.1:18789 -L 6080:127.0.0.1:6080 -L 8080:127.0.0.1:8080 ubuntu@<instance-ip>
Run openclaw onboard to configure OpenClaw, then agent-menu to select and start it. Alternatively run hermes setup, then select Hermes in agent-menu. Only one agent is selected at a time. No provider credentials are included.
OpenClaw: http://127.0.0.1:18789 (follow onboarding authentication instructions). Visual browser/noVNC: http://127.0.0.1:6080/vnc.html (SSH tunnel required; no public VNC listener). Open WebUI: http://127.0.0.1:8080 (create your first local account). Ollama API: 127.0.0.1:11434. Pull a model appropriate for available RAM with ollama pull <model>. No models are preloaded.
Helpers: agent-menu, openclaw-ai-status, setup-tailscale. Services: systemctl --user status openclaw xvfb x11vnc novnc; systemctl status ollama open-webui. Caddy is installed but disabled until you configure your own domain and HTTPS. Tailscale requires your own authentication. Keep application services loopback-only; do not expose them publicly without authentication. Documentation: https://docs.openclaw.ai
Support
Vendor support
For AMI-specific issues, contact the publisher through AWS Marketplace. Upstream components are open source and supported by their respective communities.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.