About HotCodePush
HotCodePush’s hosted MCP server lets teams managing React Native, Expo, Capacitor, and Cordova apps control OTA releases and inspect update data. It supports release operations and documentation search, while bundle uploads remain with the CLI.
Key features
- 01Release, roll out, pause, and roll back updates
- 02Manage apps and channels
- 03Read OTA update statistics
- 04Search HotCodePush documentation
- 05Enforce signed-in user permissions
Use cases
- 01Pause a production channel release
- 02Check devices that applied updates
- 03Roll back an OTA release
- 04Find HotCodePush setup guidance
MCP support
The hosted server exposes HotCodePush release management, app and channel operations, statistics, and documentation search to MCP clients over streamable HTTP. OAuth is used for signed-in tools; docs tools can be used without sign-in.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.hotcodepush.com/mcpclaude mcp add --transport http com-hotcodepush https://mcp.hotcodepush.com/mcp{"mcpServers":{"com-hotcodepush":{"type":"http","url":"https://mcp.hotcodepush.com/mcp"}}}Open in Cursor{"servers":{"com-hotcodepush":{"type":"http","url":"https://mcp.hotcodepush.com/mcp"}}}Open in VS Code{"mcpServers":{"com-hotcodepush":{"type":"http","url":"https://mcp.hotcodepush.com/mcp"}}}{"mcpServers":{"com-hotcodepush":{"serverUrl":"https://mcp.hotcodepush.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMEHotCodePush MCP Server
The official MCP server for HotCodePush, the over-the-air (OTA) update service for React Native, Expo, Capacitor and Cordova. It lets AI agents release, roll out, pause and roll back OTA updates, manage apps and channels, read update statistics and search the docs. Every tool does exactly what the signed-in person may do.
Endpoint
The server is hosted and ready to use:
https://mcp.hotcodepush.com/mcp
It speaks the streamable HTTP transport of the Model Context Protocol. The setups below sign in through OAuth. The docs tools need no sign-in: add ?toolsets=docs to the URL.
Uploading a bundle stays with the CLI. The server releases what was uploaded.
Installation
ChatGPT
ChatGPT connects a custom MCP server on a Business, Enterprise or Edu workspace, and on Pro for reads alone. Open Plugins, select the plus button, then Add custom MCP server. Enter the name HotCodePush and the URL, choose OAuth as the authentication and select Create as a plugin. Install the plugin, then type @ in a new conversation and pick it.
Claude Code
Run this in the project, or with --scope user for every project:
claude mcp add --transport http hotcodepush "https://mcp.hotcodepush.com/mcp"
Then run /mcp in Claude Code and choose hotcodepush to sign in.
Claude Desktop
Open Customize › Connectors › Add custom connector. Enter the name HotCodePush and the URL and connect. The same connector works on claude.ai.
Codex
Add the server to ~/.codex/config.toml:
[mcp_servers.hotcodepush]
url = "https://mcp.hotcodepush.com/mcp"
Then run codex mcp login hotcodepush.
Cursor
Add .cursor/mcp.json to your project, or ~/.cursor/mcp.json for every project:
{
"mcpServers": {
"hotcodepush": {
"url": "https://mcp.hotcodepush.com/mcp"
}
}
}
VS Code
Add .vscode/mcp.json to your project:
{
"servers": {
"hotcodepush": {
"type": "http",
"url": "https://mcp.hotcodepush.com/mcp"
}
}
}
Each client opens the consent page in your browser once. API tokens for headless clients, toolsets, read-only mode and other clients are covered in the docs.
Usage
Ask your agent, for example:
- "Which channels does my app have?"
- "Pause the current release in the production channel."
- "How many devices applied updates in the last seven days?"
Documentation
- MCP server: sign-in, client setup, toolsets and troubleshooting.
- Tools: every tool by toolset.
License
MIT. See LICENSE.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.