MCP.so
ログイン

Model Context Protocol (MCP) Server

@infinyte

Model Context Protocol (MCP) Server について

simple mcp server

基本情報

カテゴリ

その他

ランタイム

node

トランスポート

stdio

公開者

infinyte

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "mcp-server-infinyte": {
      "command": "docker",
      "args": [
        "compose",
        "up",
        "-d"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Model Context Protocol (MCP) Server?

A server that provides a unified API for multiple AI model providers (Anthropic, OpenAI), supporting chat completions, legacy completions, tool calling, and context handling. It is built for developers integrating AI models with persistent state management via MongoDB.

How to use Model Context Protocol (MCP) Server?

Clone the repository, run npm install, then npm run setup to configure API keys. Start the server with npm start; it will be available at http://localhost:3000. Use POST /mcp/:provider to make AI model requests and GET /tools/available to discover tools.

Key features of Model Context Protocol (MCP) Server

  • Unified API for Anthropic and OpenAI providers
  • Chat completions and legacy completions support
  • Tool calling with automatic execution
  • MongoDB persistence and state management
  • Environment-based configuration via .env file
  • Tool execution history and analytics tracking

Use cases of Model Context Protocol (MCP) Server

  • Build AI-powered chatbots with multiple model backends
  • Incorporate web search and content retrieval into AI responses
  • Generate and edit images using OpenAI or Stability AI
  • Enable AI agents to dynamically discover and call available tools

FAQ from Model Context Protocol (MCP) Server

Which AI providers are supported?

Anthropic (for Claude models) and OpenAI (for GPT models and DALL-E image generation) are supported.

How do I set up MongoDB?

MongoDB can be set up automatically with Docker via npm run setup-mongodb, manually with Docker Compose, installed locally, or used through MongoDB Atlas for production.

What tools come built-in?

Web search, web content retrieval, web batch retrieval, image generation, image editing, and image variation tools are included.

How do I start the server in production?

Use npm run pm2:start:prod to run the server with PM2 process manager for automatic restarts and production optimization.

What API keys are required?

You need keys for Anthropic, OpenAI, Stability AI (image generation), and Google CSE or Bing (web search) as appropriate for your use case.

コメント

「その他」の他のコンテンツ