mcp-server
If you are the rightful owner of mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
MCP-Server is a server and client implementation based on the Model Context Protocol (MCP), enabling large language models (LLM) to invoke external tools through a structured protocol to accomplish complex tasks.
MCP-Server is a comprehensive implementation of the Model Context Protocol (MCP) designed to facilitate the interaction between large language models (LLM) and external tools. The project consists of two main components: the MCP server and a TypeScript-based client. The server acts as a provider of tools, responsible for registering and executing various utilities such as weather forecasting and GitHub user information retrieval. The client connects to the LLM API and the MCP server, handling user queries and coordinating model and tool invocations. It offers a command-line interface and a web API based on Express, supporting streaming output and process visualization. The system architecture includes a frontend application, MCP client, and MCP server, with communication facilitated through standard input/output. The project leverages Node.js, TypeScript, and the Deepseek LLM via an OpenAI-compatible API, utilizing the Model Context Protocol for communication.
Features
- Tool Registration: Implements tools for weather forecasting and GitHub user information retrieval.
- Tool Execution: Receives client requests, executes tools, and returns results.
- Communication: Utilizes standard input/output (stdio) for client-server communication.
- Web API: Provides HTTP endpoints for frontend applications, supporting streaming output and process visualization.
- White-box Process Visualization: Displays intent recognition, tool selection, tool invocation parameters, and final answer generation.
