A service that exposes tools to AI models via the Model Context Protocol.
An MCP server is a service that exposes a set of tools to AI models via the Model Context Protocol. The server declares what tools it has (names, descriptions, input schemas), handles tool calls from AI clients, and returns structured results.
There are two types of MCP server: local (runs as a process on the same machine as the AI tool, communicates over stdio) and remote (runs as a web service, communicates over HTTP). Remote servers are more flexible — they can be shared across machines, require no local installation, and work over SSH.
A well-designed MCP server focuses on doing one thing well. It describes its tools clearly so AI models can understand when and how to use them. The tool descriptions are critical — they are what the AI reads to decide which tool to call.
The MCP ecosystem is growing rapidly. Popular MCP servers include filesystem access, web search, GitHub, Slack, databases, and time tracking tools like ClockMe.
In ClockMe
ClockMe is a remote MCP server. It runs at clockme.co/api/mcp and exposes 18 tools covering timer control (start, stop, update), project management (list, get, create), task tracking, time entry editing, reporting, and budget monitoring. Being remote means you add one URL to your AI tool — no local process, no port management.
Try ClockMe free →A local MCP server runs as a process on your computer and communicates over stdin/stdout. A remote MCP server runs as a web service and communicates over HTTP. Remote servers work across machines, don't require local installation, and can serve multiple AI clients at once.
On connection, the MCP client calls the server's tools/list endpoint. The server returns a list of tool names, descriptions, and input schemas. The AI model reads these descriptions to understand when and how to use each tool.
Yes. A remote MCP server like ClockMe's works with Claude Code, Cursor, Windsurf, GitHub Copilot, Gemini CLI, and any other MCP-compatible client. Each client connects using the same URL and API key.