SuperAPI AI Directory

mcp-server

🎖️ 📇 ☁️ 🍎 🪟 🐧 - Uptime monitoring for websites, APIs, SSL certificates, domain expiry, ping and TCP/UDP ports: manage monitors and view incidents and uptime history.

Open website

About this tool

UptyBots MCP Server is an MCP (Model Context Protocol) server used to connect Claude, Cursor, and other MCP-compatible AI assistants to UptyBots availability monitoring data. Through natural language conversations, AI assistants can access the user's UptyBots monitoring information, such as checking monitoring status, viewing incident records, creating new monitors, and more.

In terms of use, this server is mainly aimed at scenarios where AI assistants are used to manage websites, APIs, SSL certificates, domain expiration, Ping, and TCP/UDP port monitoring, allowing users to query and operate monitoring data without leaving the conversation interface.

The project provides two ways to use it. One is a hosted remote server, with no installation and no API key required: after adding this URL to the client, a browser will open for the user to log in and authorize access; if there is no UptyBots account, one can be created during that step. When using Claude Code, run /mcp, select uptybots, and complete authentication. The other is the local version corresponding to this repository, which runs on the user's own machine and uses an API key (rather than browser login) for authentication, making it suitable for MCP clients that cannot connect to a remote server, as well as offline or self-contained deployment environments. The local version exposes the same 15 tools as the hosted version.

Using the local version requires the following conditions: Node.js 18 or later, a UptyBots account, and an API key obtained from Account → API Keys. No installation step is required during configuration; npx will fetch the published package on first run. Add the corresponding configuration to the MCP client configuration file, for example Claude Desktop's claude_desktop_config.json; Claude Code's ~/.claude/settings.json and Cursor's .cursor/mcp.json use the same format. After configuration is complete, restart the AI client, and the MCP tools are immediately available.

Optionally, it is also possible to run the local checkout from source instead of using the npm package, and then point the client to that checkout directory.

This server provides a total of 15 available tools. The official description mentions that the assistant can proactively offer some starting sentences, which are only wording expressions for theexisting tool functionality and do not add extra capabilities; the hosted server also provides five of them.

The project documentation provides the following related resource links: MCP Server Docs (complete setup guide and tool reference, covering the hosted server and local version), REST API Reference (the API supporting the MCP server), and Account → Connected Applications (used to view and revoke access permissions granted through the hosted server).

Regarding pricing, billing methods, the specific tool list, environment variable list, and other content, the current materials are not complete and detailed information is not yet provided.