yitam

Kết nối tri thức y học cổ truyền với công nghệ hiện đại

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

Command

claude mcp add --transport stdio hadv-yitam node server/index.js \
  --env MCP_DOMAIN="your-domain-or-localhost" \
  --env MCP_ENABLED="true" \
  --env MCP_LOG_LEVEL="info" \
  --env MCP_SERVICE_NAME="yitam" \
  --env ANTHROPIC_API_KEY="your-anthropic-api-key"

How to use

This MCP server powers a TypeScript-based Claude chat bot with integrated Model Context Protocol (MCP) capabilities. The server acts as the backend for a chat interface, connecting to Claude via the Anthropic SDK and exposing real-time messaging through Socket.IO. The MCP integration enables tool use and dynamic tool-context enhancement during conversations, allowing the AI to access external tools and contexts as defined by MCP configurations. You’ll typically interact with the web client to chat, while the server handles authentication, API calls to Claude, and MCP-enabled tool orchestration behind the scenes.

How to install

Prerequisites:

Node.js (LTS) and npm installed
Docker and Docker Compose (for production deployments, optional in development)
An Anthropic API key
MCP integration setup (if you plan to use MCP tools)

Local development setup:

Clone the repository
Install dependencies and set up scripts (as described in the repo):
- npm run install:all
Create environment file for server:
- Copy server/.env.example to server/.env
- Add your Anthropic API key to server/.env
- Configure MCP-related variables in server/.env
Start in development mode (server and client):
- npm run dev

Production deployment (Docker):

Clone the repository and set environment variables:
- Copy server/.env.example to server/.env
- Add your Anthropic API key to server/.env
- Configure MCP-related environment variables
Build and start with Docker Compose: docker-compose up --build -d
Access the app:
- Web interface: http://localhost
- API endpoint: http://localhost/api

Notes:

Ensure your environment variables are correctly set (Anthropic key, MCP settings) before starting.
For production, consider domain/SSL setup as described in the repository documentation.

Additional notes

The MCP integration relies on environment variables to enable tool capabilities. Ensure MCP_ENABLED is set to true and that MCP_DOMAIN points to your deployment domain.
If you encounter socket connection issues, verify that the server binds to the correct host/port and that firewall rules allow traffic on the API path.
In production, use Docker Compose with the provided ssl/get-ssl-cert.sh script if you’re using Let's Encrypt certificates and a custom domain.
Logs can be inspected with docker-compose logs -f; use docker-compose ps to check container status.
The project uses an MCP SDK; make sure your MCP-related environment variables align with your MCP provider configuration.

Related MCP Servers

pluggedin-app

The Crossroads for AI Data Exchanges. A unified, self-hostable web interface for discovering, configuring, and managing Model Context Protocol (MCP) servers—bringing together AI tools, workspaces, prompts, and logs from multiple MCP sources (Claude, Cursor, etc.) under one roof.

qdrant

MCP server for semantic search using local Qdrant vector database and OpenAI embeddings

rag

⚠️ DEPRECATED - Use https://github.com/SylphxAI/coderag instead

grok-faf

First MCP server for Grok | FAST⚡️AF • URL-based AI context • Vercel-deployed

mcp-for-database

MCP Database Console is a revolutionary web application that bridges the gap between natural language and database queries. Built with Next.js and powered by the MCP-DB Connector, it allows users to interact with databases using plain English instead of complex SQL syntax.

driflyte

The Driflyte MCP Server exposes tools that allow AI assistants to query and retrieve topic-specific knowledge from recursively crawled and indexed web pages.