Command Palette

Search for a command to run...

UnylyUnyly
Весь каталог

OmniMCP Container

БесплатноНе проверен

A containerized MCP bridge that lets local or cloud LLMs access your filesystem and perform web searches via Ollama and FastMCP.

GitHubEmbed

Описание

A containerized MCP bridge that lets local or cloud LLMs access your filesystem and perform web searches via Ollama and FastMCP.

README

Python License

Docker Process FastMCP MCP on IDE

A containerized Model Context Protocol (MCP) bridge for local and external LLMs. This project provides a robust, extensible hub that allows you to use your local models (via Ollama) or cloud models while giving them access to your local filesystem and web search tools.


web_search mocked fetch_private_data using api.example.com


🌟 Key Features

  • Local Inference: Integrated Ollama service for running models locally (no internet required).
  • Bridge Service: Python-based MCP server using FastMCP that acts as the tool hub.
  • Filesystem Access: Secure access to your project files via mounted Docker volumes.
  • Web Search: Extensible web search tool for real-time information.
  • SSE Transport: Exposes the MCP server over HTTP (SSE) for easy integration with other apps.

🛠️ Stack

  • Orchestration: Docker Compose
  • Inference: Ollama
  • MCP Framework: FastMCP (Python)
  • Networking: LAN-ready SSE (Server-Sent Events)

🚀 Getting Started

1. Configure Environment

Copy the template and fill in your API keys (optional for local-only use):

cp .env.template .env

2. Launch the Hub

docker-compose up -d --build

3. Verify Services

  • Ollama: Accessible at http://localhost:11434
  • MCP Bridge: Accessible at http://localhost:8000/sse

📂 Mounting Your Project

To give the LLM access to your specific project files, place them in the my_project_files/ directory, or modify the volumes section in docker-compose.yml:

volumes:
  - ./your_local_path:/data

🔗 Connecting to Claude Desktop

To use this as an MCP server in Claude Desktop, add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "omnimcp": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/inspector", "http://localhost:8000/sse"]
    }
  }
}

Note: This utilizes the MCP inspector to bridge the SSE connection to Claude Desktop's stdio-based expectations, or you can use other clients that support SSE directly.

🛠️ Customizing Tools

Add new tools in app.py by using the @mcp.tool() decorator.

📄 Licencia

Este proyecto está bajo la Licencia MIT - mira el archivo LICENSE para detalles.

from github.com/hectorflores28/MCP-Template

Установка OmniMCP Container

У этого сервера нет опубликованного пакета — он собирается из исходников. Открой репозиторий и следуй инструкции в README.

▸ github.com/hectorflores28/MCP-Template

FAQ

OmniMCP Container MCP бесплатный?

Да, OmniMCP Container MCP бесплатный — установка в пару кликов через Unyly без оплаты.

Нужен ли API-ключ для OmniMCP Container?

Нет, OmniMCP Container работает без API-ключей и переменных окружения.

OmniMCP Container — hosted или self-hosted?

Self-hosted: сервер запускается локально на твоей машине командой из раздела установки.

Как установить OmniMCP Container в Claude Desktop, Claude Code или Cursor?

Открой OmniMCP Container на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.

Похожие MCP

Compare OmniMCP Container with

Не уверен что выбрать?

Найди свой стек за 60 секунд

Автор?

Embed-бейдж для README

Похожее

Все в категории development