{"slug": "show-hn-openclawmachines-extending-openclaw-to-the-enterprise", "title": "Show HN: OpenClawMachines – Extending OpenClaw to the Enterprise", "summary": "OpenClawMachines, an open-source platform for running isolated OpenClaw agents in Firecracker microVMs on self-owned hardware, was released. The platform provides a control plane, host agent, LLM proxy, and browser runtime for secure AI sandboxing with hardware isolation and edge auth. It aims to offer cost-effective, sovereign AI infrastructure for enterprises.", "body_md": "**Run as many isolated OpenClaw agents as you need, on hardware you own.**\n\nOpenClaw Machines is an **open-source platform for running\nOpenClaw in secure AI sandboxes on your\nown infrastructure**. A\n\n**control plane** orchestrates your hosts, and each agent runs in its own\n\n[Firecracker](https://github.com/firecracker-microvm/firecracker)microVM on them — hardware-isolated, safe for untrusted and agent-generated code. A\n\n**Cloudflare data plane** is the front door: every machine gets its own subdomain behind edge auth, reached through a tunnel that terminates\n\n*inside*the VM — no host port is exposed for user-to-VM traffic. The current control plane still needs private or firewall-restricted access to each agent's authenticated control API on\n\n`9090`\n\n. See it running at\n[openclawmachines.com](https://openclawmachines.com).\n\nThe Apache-2.0 **public core** ships every piece of that stack:\n\n- a\n— Go API, Postgres-backed accounts, machines, and hosts;**minimal control plane**[placement](/mathaix/OpenClawMachines/blob/main/docs/architecture.md#scheduling--capacity-management),[machine lifecycle](/mathaix/OpenClawMachines/blob/main/docs/architecture.md#machine-lifecycle),[host enrollment](/mathaix/OpenClawMachines/blob/main/docs/host-enrollment.md),[backups](/mathaix/OpenClawMachines/blob/main/docs/architecture.md#backup--restore), and[durable workflows](/mathaix/OpenClawMachines/blob/main/docs/architecture.md#durable-workflows-dbos); - the\n(**host agent**`ocm-agent`\n\n) — boots, supervises, and reaps Firecracker microVMs on your enrolled Linux boxes, managing bridge/TAP networking and rootfs staging; - a per-host\n(LiteLLM) — one place for model keys and BYO-key support, with per-machine usage tracking across providers (or your own locally served models);**LLM proxy** - the\n— the in-VM pieces: auth proxy, web-chat gateway, live terminal, and the artifact-driven runtime staging/upgrade flow;**OpenClaw runtime** - the\n— paired Chromium browser VMs with CDP routing and a watchable live view;**browser runtime** — GitHub, Google Workspace, OpenAPI, GraphQL, and remote-MCP tools connected once per workspace and exposed to machines through the OCM MCP facade;**workspace integrations / native MCP**- and the\nthat assemble it all — every component's build command, the GCS artifact bucket layout, host provisioning scripts, and the**build pipelines**[release lanes](/mathaix/OpenClawMachines/blob/main/docs/ci-release.md).\n\nThe `ocm`\n\nCLI lives in the separate\n[ mathaix/ocm-cli](https://github.com/mathaix/ocm-cli) Apache-2.0 repository.\n\nClick the screenshot to watch the 43-second demo on YouTube. This is a linked image, not an embedded player.\n\nThe demo covers host onboarding, agent spin-up, the running Firecracker VM terminal, workspace MCP integrations, and an agent tool call end to end.\n\n**Security.** Real isolation, not containers: one Firecracker microVM per agent, with its own guest kernel behind a KVM hardware boundary — and auth enforced at the edge and again inside every VM.**Cost.** One flat server cost: rent a single bare-metal box and run as many hardware-isolated agents as it fits — see[how the options compare](#how-the-options-compare). The same architecture cuts token spend too: route agents to open-source models running on your own GPU hardware instead of paying per-token APIs.**Sovereignty.** Your hardware, your data, your keys. Run the control plane and workers on machines you own, and route model traffic through the per-host LLM proxy to any provider — or to models served on your own GPUs.**Open source.** Apache-2.0 public core and companion, permissively licensed for adoption, embedding, and contribution.`ocm`\n\nCLI**Enterprise.** Multi-user accounts and teams, admin-gated host management, encrypted per-machine secrets, and capacity/placement policies across your fleet.**Ecosystem.** Browser VMs for web automation, live terminal and web chat, per-VM routing, workspace-scoped native MCP integrations, backups/snapshots, agent memory, and observability with OpenTelemetry/Opik tracing and per-machine usage tracking.\n\nIf you run [OpenClaw](https://github.com/openclaw/openclaw) today, you have a\nfew options:\n\n**Local hardware**— run it on your own laptop or desktop.** A VPS**(e.g. Hostinger, DigitalOcean) — rent a virtual server and run it there.** A managed service**(e.g. KiloClaw) — spin up a hosted OpenClaw instance and pay per instance.\n\nOpenClaw Machines is the **fourth option**: rent **one bare-metal server**\n(OVHcloud, Hetzner, …), point OpenClaw Machines at it, and spin up as many\n**hardware-isolated** OpenClaw instances as the box will hold. One agent or\nfifty — **the cost stays one flat server**.\n\nIn short: the **managed** route is easiest but priced per agent; **local** and\n**VPS** are cheap to start but don't isolate or scale well. **OpenClaw Machines**\ntrades a little more setup for the best economics and isolation once you're\nrunning more than a couple of agents — one server, many hardware-isolated agents,\nall yours.\n\nOpenClaw Machines turns your own Linux servers into a pool of secure, on-demand\nsandboxes. Each sandbox is a real Firecracker microVM (its own kernel,\nhardware-isolated via KVM) that runs one AI agent. The platform is the **control\nplane** that creates those VMs, keeps track of them, routes traffic to them, and\ntears them down — so you can run many untrusted agents safely on infrastructure\nyou own. Think: a mini-cloud for AI agents, that you self-host.\n\n**Control plane (Go backend) — the brain.** Accounts, machines, hosts, and config; the API the UI/CLI call; placement and lifecycle orchestration.**Hosts + worker agents — your Linux boxes.** Enroll a host with an install script; its worker agent boots and stops Firecracker microVMs when told to.**Machines — one isolated microVM per agent.** Inside: the OpenClaw agent, a web chat gateway, and a live terminal.**Browser VMs**— separate microVMs running headful Chromium with a live view, driven by the agent over CDP for browser automation.** Routing / data plane**— every running VM gets its own subdomain and a Cloudflare Tunnel that terminates** inside the VM**, with auth enforced at the edge and again in-VM.** Workspace integrations (native MCP)**— connect external tools once per workspace (GitHub, Google Workspace, or any OpenAPI / GraphQL / remote-MCP endpoint); the control plane exposes them to each machine's agent through a single built-in MCP server, so the agent discovers and calls them with`ocm.search_tools`\n\n/`ocm.call_tool`\n\ninstead of per-integration wiring.\n\n``` php\nflowchart TB\n    U[\"you — browser / ocm CLI\"] --> EDGE[\"Cloudflare edge<br/>Access auth · Worker route lookup (KV)\"]\n    EDGE -->|dashboard / API| CP[\"Control plane (Go)<br/>accounts · machines · hosts<br/>placement · lifecycle · backups\"]\n    CP --- DB[(\"Postgres\")]\n    CP -->|enroll · heartbeat · boot/stop :9090| H1[\"Host 1 — your Linux box<br/>ocm-agent · LLM proxy · CDP proxy\"]\n    CP -->|…| HN[\"Host N\"]\n    EDGE -->|per-VM tunnel, terminates inside the VM| VM1\n    subgraph H1X[\"Host 1's microVMs\"]\n        VM1[\"Machine — Firecracker microVM<br/>OpenClaw agent · web chat · terminal<br/>authproxy + cloudflared inside\"]\n        BVM[\"Browser VM<br/>headful Chromium · live view\"]\n        VM1 -->|CDP| BVM\n    end\n    H1 --- H1X\n```\n\nThe full design — data plane, routing, tunnels, lifecycle, config, and the\nbuild/release flow — is in ** docs/architecture.md**, and\nthe five-layer stack (React UI → Cloudflare edge → Go control plane → host\nagents → Firecracker sandboxes) is in\n\n**.**\n\n[docs/tech-stack.md](/mathaix/OpenClawMachines/blob/main/docs/tech-stack.md)OpenClaw Machines runs Firecracker microVMs, which require KVM. You need a KVM-enabled Linux host: bare metal, or a cloud VM with nested virtualization enabled. It does not run on macOS, Windows/WSL, or a standard cloud VM without nested virtualization.\n\nCheck your host:\n\n```\nmake preflight\n```\n\n** The Getting Started guide** is three stages, each\nending with something working:\n\nUsing a coding agent?Point it at[and ask it to follow the guide from Stage 1.]`docs/getting-started.md`\n\n**Local evaluation**— the full stack + a real Firecracker machine on one KVM-capable Linux box. No Cloudflare or public domain is required; use an existing KVM host or the optional GCP provisioning example.**Cloudflare + a dedicated host**— the production-shaped deployment: domain, tunnels, edge auth, and an enrolled cloud or bare-metal host.** The full workflow**— create and use machines (chat, terminal, browser VMs), lifecycle, backups, runtime upgrades.\n\n— the three-stage guide above**Getting Started**— using a machine day-to-day (model, chat, terminal, browser VM, files, logs, traces, backups)** User guide**— connect GitHub, Google Workspace, OpenAPI, GraphQL, and remote-MCP tools once per workspace** Workspace integrations / native MCP**[Architecture](/mathaix/OpenClawMachines/blob/main/docs/architecture.md)— data plane, routing, tunnels, lifecycle,[workspace integrations / native MCP](/mathaix/OpenClawMachines/blob/main/docs/architecture.md#workspace-integrations--native-mcp)[Tech stack](/mathaix/OpenClawMachines/blob/main/docs/tech-stack.md)— the five layers, client to sandbox[Local and BYO-host setup](/mathaix/OpenClawMachines/blob/main/docs/local-setup.md)[Control plane deployment profiles](/mathaix/OpenClawMachines/blob/main/docs/control-plane-profiles.md)[Self-hosted control plane prerequisites](/mathaix/OpenClawMachines/blob/main/docs/self-hosted-control-plane.md)[LLM operator runbook](/mathaix/OpenClawMachines/blob/main/llms/self-hosted-setup.txt)[Public docs inventory](/mathaix/OpenClawMachines/blob/main/docs/public-docs-inventory.md)[Contributing](/mathaix/OpenClawMachines/blob/main/CONTRIBUTING.md)·[Security policy](/mathaix/OpenClawMachines/blob/main/SECURITY.md)·[Code of conduct](/mathaix/OpenClawMachines/blob/main/CODE_OF_CONDUCT.md)`ocm`\n\nCLI project\n\n— questions, ideas, show & tell[GitHub Discussions](https://github.com/mathaix/OpenClawMachines/discussions)— bugs and feature requests[Issues](https://github.com/mathaix/OpenClawMachines/issues)— the open-source readiness tracker: what's done, what's next[Roadmap](https://github.com/mathaix/OpenClawMachines/issues/25)- Found a vulnerability? See the\n[security policy](/mathaix/OpenClawMachines/blob/main/SECURITY.md).\n\nSee [CONTRIBUTING.md](/mathaix/OpenClawMachines/blob/main/CONTRIBUTING.md) and the\n[code of conduct](/mathaix/OpenClawMachines/blob/main/CODE_OF_CONDUCT.md).", "url": "https://wpnews.pro/news/show-hn-openclawmachines-extending-openclaw-to-the-enterprise", "canonical_source": "https://github.com/mathaix/OpenClawMachines", "published_at": "2026-07-13 17:45:29+00:00", "updated_at": "2026-07-13 18:05:41.824364+00:00", "lang": "en", "topics": ["ai-infrastructure", "ai-agents", "ai-safety"], "entities": ["OpenClawMachines", "Firecracker", "Cloudflare", "LiteLLM", "GitHub", "Google Workspace", "OpenAPI", "GraphQL"], "alternates": {"html": "https://wpnews.pro/news/show-hn-openclawmachines-extending-openclaw-to-the-enterprise", "markdown": "https://wpnews.pro/news/show-hn-openclawmachines-extending-openclaw-to-the-enterprise.md", "text": "https://wpnews.pro/news/show-hn-openclawmachines-extending-openclaw-to-the-enterprise.txt", "jsonld": "https://wpnews.pro/news/show-hn-openclawmachines-extending-openclaw-to-the-enterprise.jsonld"}}