CloudCLI

What is CloudCLI?

CloudCLI is a cloud development environment purpose-built for AI coding agents. Claude Code, Cursor CLI, OpenAI Codex, and Gemini CLI come pre-installed in isolated containers that persist even when you close your laptop or walk away from your desk. You can start a session from your phone, browser, or via API and automation tools like n8n, Linear, or Jira, then continue it in VS Code or Cursor over SSH. Teams can share environments, MCP servers, configurations, and context files across projects. The open source project (ClaudeCodeUI) has over 8,500 GitHub stars and 1,100 forks, and the hosted

How is CloudCLI different from Claude Code Remote Control?

Claude Code Remote Control lets you view a local session from your phone, but your laptop has to stay open and connected. Close the lid or lose network for about 10 minutes and the session dies. It also only exposes the single active session, only works with Claude Code (not Cursor or Codex), and cannot start new sessions from mobile. CloudCLI runs your agent in a persistent cloud environment so nothing depends on your local machine. You get support for multiple agents, a full UI with file explorer and Git integration, access to all your sessions at once, and the ability to kick off new work f

How is CloudCLI different from Claude Code on the web?

Claude Code on the web runs sessions on Anthropic's managed cloud infrastructure and only supports Claude Code. Each session starts fresh without access to your local toolchains or custom setup. CloudCLI gives you your own persistent environment where you can run Claude Code alongside Cursor CLI, Codex, and Gemini CLI in the same workspace. You get SSH access to connect from VS Code or any desktop IDE, a full file explorer and Git UI, and the ability to install your own dependencies, databases, and tools. Your environment persists between sessions so you never start from scratch.

How is CloudCLI different from GitHub Codespaces?

GitHub Codespaces provides general-purpose cloud dev environments, but it is not built for AI coding agents. CloudCLI comes with Claude Code, Cursor CLI, Codex, and Gemini CLI pre-installed and includes a mobile-optimized UI designed for interacting with AI agents on the go. It also works with GitLab and Bitbucket, not just GitHub, and is fully open source so you can self-host it. Pricing starts at $7/month compared to Codespaces' per-hour billing model. See the full CloudCLI vs GitHub Codespaces comparison.

Do I need my own AI subscription?

Yes. CloudCLI provides the cloud environment and the tools, not the AI itself. You bring your own API key or subscription for Claude Code, Cursor, Codex, or Gemini. This means you are not locked into a single provider and can switch agents or run multiple ones side by side. Your keys are stored encrypted in your environment and never shared.

What happens when I close my laptop?

Nothing. Your environment keeps running in the cloud. Terminal sessions, running processes, and file changes all persist. You can close your laptop, commute, sleep, and pick up the exact same session from your phone or another computer. This is the core difference between CloudCLI and tools like Remote Control or the SSH + tmux + Tailscale DIY stack that depend on your local machine staying awake. You can even trigger background coding agents from your phone or via API and come back to finished code.

Can I self-host CloudCLI instead of using the cloud version?

Yes. CloudCLI is built on ClaudeCodeUI, which is fully open source under GPL-3 with over 8,500 GitHub stars. Run it locally with a single npx command, deploy it on your own VPS, or set it up on your team's infrastructure. The hosted version at cloudcli.ai adds managed cloud environments, persistent sessions, cross-device access, and team features on top of the same open source core. Check the self-hosting guide.

Can my team share environments and configurations?

Yes. Teams can share MCP servers, context files like CLAUDE.md and .cursorrules, environment variables, and tool configurations across projects. Each developer still gets their own isolated container, but everyone starts from the same setup. This eliminates the 'works on my machine' problem and means new developers can onboard in minutes instead of hours. Learn more about Claude Code for teams.