
MCP Hospital Assistant
An intelligent hospital appointment management system that allows users to add doctors/patients, book appointments, and view medical schedules through both a manual UI and an LLM-powered assistant interface.
README
🏥 MCP Hospital Assistant
A full-stack intelligent hospital appointment management system powered by React, Node.js, PostgreSQL, and Mistral LLM (via Ollama).
🔥 Preview
🎥 Demo Video
▶️ Click here to watch the demo
🏠 Dashboard
💬 LLM Chat Assistant
💬 Mannual Booking
🔁 Node.js + Express
🎨 React.js (frontend)
🧠 LLM (Ollama - Mistral)
💾 PostgreSQL
🛡️ Session-based login
📷 UI Preview
Login / Register Page
Dashboard with buttons to manage system manually
Book Appointment (manual)
LLM-powered auto assistant interface
📁 Project Structure
├── backend/ │ ├── server.js # Node.js + Express backend │ ├── config/config.js # PostgreSQL config │ ├── mcp/ # All business logic handlers │ ├── auth/ # Login, register, session routes ├── frontend/ │ └── mcp-ui/ # React frontend │ ├── components/ # AddDoctor, Chat, Login, Home etc. │ └── App.js # Main route controller
⚙️ Setup Instructions
- ✅ Install PostgreSQL and create DB
sudo apt install postgresql postgresql-contrib sudo -u postgres psql
Then run:
CREATE DATABASE mcp; CREATE USER mcpuser WITH PASSWORD 'mcppass'; GRANT ALL PRIVILEGES ON DATABASE mcp TO mcpuser;
- 📦 Create Schema
Connect:
psql -U mcpuser -d mcp -h localhost
CREATE TABLE doctors ( doctor_id SERIAL PRIMARY KEY, name TEXT, department TEXT, available_slots JSONB );
CREATE TABLE patients ( patient_id SERIAL PRIMARY KEY, name TEXT, age INTEGER, contact TEXT );
CREATE TABLE appointments ( appointment_id SERIAL PRIMARY KEY, doctor_id INTEGER REFERENCES doctors(doctor_id), patient_id INTEGER REFERENCES patients(patient_id), appointment_time TEXT );
CREATE TABLE users ( id SERIAL PRIMARY KEY, username TEXT UNIQUE, password TEXT );
GRANT ALL PRIVILEGES ON ALL TABLES IN SCHEMA public TO mcpuser; GRANT ALL PRIVILEGES ON ALL SEQUENCES IN SCHEMA public TO mcpuser;
- 🧠 Install Ollama with Mistral
curl -fsSL https://ollama.com/install.sh | sh ollama run mistral
Make sure the local Ollama server is running and available on default port 11434. 4. 🚀 Backend Setup (Node.js)
cd backend/ npm install node server.js
This starts backend on: http://localhost:3007 5. 💻 Frontend Setup (React)
cd frontend/mcp-ui npm install npm start
This starts frontend at: http://localhost:3000 🧠 Key Functionalities ✨ Mode 1: Manual via UI
You can click on homepage icons to:
➕ Add Doctor
➕ Add Patient
📅 Book Appointment
📋 View Doctors
📋 View Appointments
These UI features directly use API endpoints (/api/execute, /api/book, etc.) 🔍 Mode 2: Smart Assistant (LLM-enabled)
From the "Ask Assistant" chat interface:
Type: Show all appointments
Or: Book appointment for patient John with doctor 2 at 10:00
✅ If MCP Tools is enabled (checkbox at top), the LLM will:
🔎 Understand user command
🔧 Choose correct tool from tools.json
🧠 Generate SQL
🔁 Automatically execute API calls behind the scenes
❌ If disabled, LLM simply gives plain answers (no system change). 🔐 Authentication
Only authenticated users can access core pages.
Register page stores hashed credentials securely in PostgreSQL.
Session stored using express-session.
📝 Example Prompts for LLM
Show all the doctors in doctor table Show appointments after 11:00 Book an appointment with Dr. Renu at 09:00 for John (age 30)
✅ Summary Feature Manual UI LLM Tools Add Doctor ✅ Yes ✅ via prompt Book Appointment ✅ Yes ✅ if slot available View Doctors/Appointments ✅ Yes ✅ via prompt Edit Slots ✅ Yes (via query) ✅ via prompt Security 🔐 Session Login 🔐 Session Protected
推荐服务器

Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。