English Français العربية

CloudCrane

Official registry Productivity

Ask your assistant for an app and it lands on your phone: offline, data on the phone.

Source code Website Documentation

Install

Add this to your client's MCP configuration (Claude Desktop: claude_desktop_config.json; Cursor: .cursor/mcp.json; VS Code and others use the same "mcpServers" shape).

Remote (streamable-http)
Claude Code 
claude mcp add --transport http cloudcrane 'https://cloudcrane.ai/api/build/mcp'
JSON configuration (Claude Desktop, Cursor, VS Code) 
{
    "mcpServers": {
        "cloudcrane": {
            "type": "http",
            "url": "https://cloudcrane.ai/api/build/mcp"
        }
    }
}

Documentation

From the project's README on GitHub, refreshed weekly. Links open on GitHub.

Contents
  1. CloudCrane MCP
  2. Connect
  3. Tools
  4. Try
  5. Things that look like a broken server
  6. Links

CloudCrane MCP

Ask your assistant for an app, and it lands on your phone's home screen.

Your assistant writes a small app for you (a habit tracker, a spending log, a packing checklist, a workout timer) and publishes it with CloudCrane. You get a link. Open it on your phone, add it to your home screen, and it's an app of its own: its own icon, full screen, working offline.

  • Your data stays on your phone. What you save in an app is kept on the phone, never on our servers.
  • Change it by asking. "Add a reading habit", "make the buttons bigger": your assistant updates the app, and the new version arrives the next time you open it, with everything you saved still there.
  • Every version is kept. Go back to any earlier version, or delete an app for good.
  • Each app is locked down. Every app runs at its own address, can't talk to other websites, and is checked before it's published: no password or card fields, no forms that send your entries anywhere.

Connect

One remote MCP server, Streamable HTTP (stateless, no SSE stream), with OAuth sign-in and dynamic client registration. Nothing to install.

https://cloudcrane.ai/api/build/mcp

(https://cloudcrane.ai/api/apps/mcp is the same server.)

Add the URL with no key. The client opens a CloudCrane page where you sign in (Google or an emailed code), pick your workspace and choose what the assistant may do: Read only, or Read and build, which also lets it make and change your apps. You can sign an assistant out at any time from My apps in the dashboard.

Sign-in has been tested with Claude, ChatGPT, Cursor, Cline and Smithery. Exact settings for each are in llms-install.md, which an agent can follow to set it up.

Claude: Settings → Connectors → Add custom connector with the URL above.

ChatGPT: add a custom MCP server with the URL above and choose OAuth.

Cursor (~/.cursor/mcp.json):

{ "mcpServers": { "cloudcrane": { "url": "https://cloudcrane.ai/api/build/mcp" } } }

Cline (cline_mcp_settings.json):

{ "mcpServers": { "cloudcrane": { "transport": { "type": "streamableHttp", "url": "https://cloudcrane.ai/api/build/mcp" } } } }

Tools

ToolWhat it doesNeeds
list_appsYour apps, with their install linksRead
get_appAn app's code, to read or change itRead
list_versionsEvery version of an app, and which one is installedRead
publish_appPublishes a new app and returns its install linkBuild
update_appPublishes a new version of an appBuild
rollback_appServes an earlier version againBuild
delete_appDeletes an app and every version of itBuild

The server tells the assistant how to write an app that works here when it connects: one self-contained HTML file, mobile first, data kept on the phone, and saved data carried across every update. The tools that change apps are marked as such, so clients like Claude ask you before each call.

Try

  1. "Make me a habit tracker for water, reading and the gym, and put it on my phone."
  2. "Add a 'walk 20 minutes' habit to my Habits app."
  3. "My packing list app: go back to the version from before the last change."

Things that look like a broken server

  • 406: MCP requires Accept: application/json, text/event-stream on every POST, even though this server never sends a stream.
  • 405 on GET: the server is stateless and offers no SSE stream, so only POST is allowed. A client that silently falls back to SSE connects but lists no tools.
  • 403: a request from a browser (any request with an Origin header) is refused. Call it from a server or a desktop client.

This repository holds documentation and the registry entry (server.json), not the server's source.

Frequently asked questions

What is CloudCrane?

Ask your assistant for an app and it lands on your phone: offline, data on the phone.

How do I install CloudCrane?

It runs as a remote server. In Claude Code: claude mcp add --transport http cloudcrane 'https://cloudcrane.ai/api/build/mcp'. For other clients, paste the JSON configuration from this page into their MCP settings.

Does CloudCrane need an API key?

Its registry entry lists no required key or secret. Some features may still need an account with the service it connects to.

Is CloudCrane open source?

Yes. Its code is on GitHub under the MIT license.

Who maintains CloudCrane?

It is published by cloudcrane-dev. AI.tn lists it but does not maintain or endorse it.

From the official MCP Registry. Listing is not an endorsement: review a server's code and permissions before giving it access to your data.