AI Hosting
Your Agents Run Here. And They Run This Place.
xCloud is where your AI agents live, and the thing your AI agent can operate. Deploy Claude Code or an always-on assistant onto a server you own, then manage that server by describing what you want.
The whole thing, once
One Account, Two Directions.
Agents you already use reach in and operate your infrastructure. Agents you deploy live inside it. Same account, same panel, same bill.
The agent you already use
Claude Code
Cursor
Codex- Any MCP client
Your xCloud account
Servers you own · 31 locations · free SSL · patched and backed up by us
In practice
Things People Actually Type.
No endpoints, no resource IDs, no menu to find. Your agent maps the sentence to the right operation, runs the steps in order, and waits for each one to finish.
Which of my sites have plugin vulnerabilities?
xcloud:wordpressReboot staging-01 and tell me when it's back.
xcloud:serversBack up the shop before I touch the theme.
xcloud:sitesRenew the certificate on every domain expiring this month.
xcloud:sslWhat PHP version is each server on?
xcloud:serversGive me a one-click login to the staging site.
xcloud:wordpressPurge the cache on everything in production.
xcloud:sitesAdd a cron job to run the queue worker every minute.
xcloud:servers
Three doors, same account
Pick How Your Agent Gets In.
All three are free with every xCloud account and all three reach the same operations. They differ in what you have to set up, and in what is holding the credentials.
01xCloud MCPRecommended
One Endpoint, Whichever Agent You Run.
Add a single URL to Claude Code, Claude Desktop, Cursor, Codex, Windsurf, OpenCode or anything else that speaks MCP. OAuth fires the first time the agent reaches for an xCloud tool, so there is no token sitting in a config file, and you revoke access from the dashboard rather than by editing a machine.
- OAuth on first tool use. Nothing to paste, nothing to store
- Stdio-only clients connect through
npx mcp-remote - Headless and CI use a token scoped to
mcp:invoke, since no browser can finish OAuth
- Claude Code
- Cursor
- Codex
- Any MCP client
app.xcloud.host/mcpOAuth on first useclaude mcp add xcloud --transport http https://app.xcloud.host/mcpcodex mcp add xcloud --url https://app.xcloud.host/mcpxcloud:serversServers, PHP, databases, cron, firewall and fail2ban, sudo users, services, provisioningxcloud:sitesSite lifecycle: status, backups, domains, cache, SSH, git settings, deploysxcloud:wordpressPlugins, themes, updates, WP_DEBUG, magic login, vulnerabilities, PageSpeedxcloud:sslCertificates: view, install, renew, status, deletexcloud:accountCurrent user, API tokens, Cloudflare integrations, blueprints, health
claude plugin marketplace add xCloudDev/xcloud-agent-skillsclaude plugin install xcloud@xcloud-agent-skillsthen /reload-plugins to activate02AI Agent Skills
Five Skills. The Agent Picks, Not You.
The MCP hands your agent the tools; the skills teach it xCloud's workflows: which tool to reach for, the guardrails around it, how to poll an asynchronous job to completion, and how to chain a change that takes several steps. Install both.
It acts only with the permissions on the token you issued, revocable at any time. Before a bigger change it backs up first, waits for the job to actually finish, and can roll back.
Browse Agent Skills03xCloud Public API
And Underneath, It Is Just REST.
Everything the MCP exposes is a Public API operation underneath, so there is no capability gap between the paths. Use it directly when the integration lives inside your own product: a backend service, a provisioning script, a CI job, an agent loop of your own.
- Base URL
- https://app.xcloud.host/api/v1
- Auth
- Bearer token (Laravel Sanctum)
- Reference
- OpenAPI 3.0.3
curl -s https://app.xcloud.host/api/v1/health{"status":"ok","version":"v1"}No auth on /health, which is what makes it useful: it isolates a network problem from a credential one.
Full Accessread:serverswrite:serversread:siteswrite:sitesmcp:invoke
Settings → Developers → API Tokens. Give each token the narrowest scope that covers its job.
Heads up. Write operations are asynchronous. They return a task you poll, not a finished resource, so build in polling and a timeout rather than assuming the thing exists the moment the call returns.
The agents
The Whole Catalogue, In Motion.
Everything below either runs on a server you own, or connects to your account and runs it. Several do both.
Runs on your server
- Claude Code
- Codex
OpenClaw
Hermes
Paperclip- Odysseus
n8n
Ollama
LibreChat
Open WebUI
Connects to your account
- Claude Code
- Codex
- Cursor
- Antigravity
Gemini
Grok
Claude Desktop- Any MCP client
Deploy in one click, on a machine we provision, patch and monitor.
Choosing
Start From What You Want To Do.
Everything you can deploy, sorted by the job rather than the brand. Each links to its own setup steps and pricing.
| I want… | Deploy | How you reach it | From |
|---|---|---|---|
| a coding agent that keeps working after I close the laptop | Claude Code | Terminal, Claude web and mobile apps | $5.00/mo |
| to put the app Codex just built on a real domain | Codex | Your own domain, redeployed on git push | $5.00/mo |
| an assistant I can text like a person | OpenClaw | Telegram, WhatsApp, Slack, Discord | $9.99/mo |
| an agent that remembers everything between conversations | Hermes | Telegram, plus your tools, files and APIs | $9.99/mo |
| a workforce of agents with roles and budgets | Paperclip | Its own dashboard | $9.99/mo |
| a private AI workspace: chat, memory, research, documents | Odysseus | A private web workspace, your own LLM keys | $5.00/mo |
| to automate the plumbing between all of the above | n8n | 500+ integrations, webhooks, AI nodes | $5.00/mo |
| to run something that is not on this list | Your own stack | Deploy from Git, or a Docker Compose file | $3.00/server/mo |
Odysseus deploys from a Docker Compose file rather than as a one-click app. You add the compose file, an admin login and your own LLM API keys.
Bring Your Own Model, We Run The Server
Connect the provider you already pay for. Anthropic (API key or Claude Code CLI token), OpenAI, OpenRouter, Moonshot AI (Kimi) and Gemini are supported today; Grok, xAI and Mistral are coming.
Every one of them runs on the same machine underneath.
- Hardware
- 3GHz+ Intel XeonNVMe SSDNGINX or OpenLiteSpeed31 locations
- Security
- Free SSL, auto-renewedFirewall + fail2banAutomated patchingAI Bot Blocker
- Recovery
- Scheduled backupsAny S3-compatible storeSnapshotsOne-click restore
- Visibility
- Server + site monitoringAccess and deploy logsVulnerability scansPageSpeed history
Bring your own cloud too: xCloud Managed, or your own machine at DigitalOcean, Vultr, Linode, AWS or anywhere else.
The Questions That Decide It
What is free, which door to use, and why two rows of that table have different prices.
What is free, and what do I pay for?
The control plane is free. MCP, the AI Agent Skills and the Public API are included with every xCloud account, including the free plan, so there is no add-on and nothing to upgrade. You pay only when you run something: a server. Your AI assistant's own model usage is billed by its provider, not by us.
Should I use MCP or the Agent Skills?
MCP first. It is one endpoint, it works with every MCP-compatible client, and OAuth means there is no token to store. The Skills are for Claude Code specifically and add ready-made workflows on top; install both if you live in Claude Code.
Why do the prices in the table differ?
Because they are two different products. Claude Code, Codex, Odysseus and n8n run on a standard managed server, which starts at $5.00/mo. OpenClaw, Hermes and Paperclip run on a dedicated Cloud VPS sized for an always-on agent, which starts at $9.99/mo. Each row links to the ladder that applies to it.
Can I bring my own model, and my own server?
Both. Connect Anthropic, OpenAI, OpenRouter, Moonshot AI (Kimi) or Gemini today, with Grok, xAI and Mistral coming. And you can run any of this on an xCloud Managed Server or on your own machine at DigitalOcean, Vultr, Linode, AWS or another provider.
Start Free. Bring Your Own Model.
The free plan gets you a server, ten sites and the whole control plane (MCP, Agent Skills and the Public API), with no card.