✈️ AI Travel Planner MCP

An AI-powered Travel Planning Assistant built using FastMCP, LangGraph, LangChain, FastAPI, and NiceGUI.

This project was created while exploring Model Context Protocol (MCP), Agentic AI, and LangGraph workflows through a practical real-world use case.

The application helps users plan trips by fetching live weather information, generating packing suggestions, and providing AI-powered travel recommendations based on their destination and budget.

🚀 Features

• 🌍 Destination-based travel planning

• 🌤 Real-time weather information

• 🎒 Smart packing recommendations

• 🤖 AI-powered travel suggestions

• 🔗 MCP Tool Integration

• 🧠 LangGraph Agent Workflow

• ⚡ FastAPI Backend

• 🎨 Modern NiceGUI Interface

• 🌙 Dark Mode Support

Related MCP server: Travel MCP

🏗️ Architecture

User Input
     │
     ▼
NiceGUI Interface
     │
     ▼
FastAPI Backend
     │
     ▼
LangGraph Workflow
     │
 ┌───────────────┐
 │ Weather Agent │
 └───────┬───────┘
         │
 ┌───────▼───────┐
 │ Packing Agent │
 └───────┬───────┘
         │
 ┌───────▼──────────┐
 │ Travel Advisor   │
 └───────┬──────────┘
         │
 ┌───────▼──────────┐
 │ Final Report     │
 └───────┬──────────┘
         │
         ▼
Travel Recommendation

🧠 MCP Tools

Location Tool

Uses OpenStreetMap's Nominatim API to retrieve geographical coordinates from a destination name.

Weather Tool

Uses Open-Meteo API to fetch real-time weather information.

Packing Tool

Generates packing suggestions based on weather conditions.

🛠️ Tech Stack

AI & Agents

• LangChain

• LangGraph

• FastMCP

• Groq LLM

Backend

• FastAPI

• Python

Frontend

• NiceGUI

APIs

• Open-Meteo API

• OpenStreetMap Nominatim API

📂 Project Structure

travel-planner-mcp/

├── app.py
├── graph.py
├── state.py
│
├── agents/
│   ├── weather_agent.py
│   ├── packing_agent.py
│   ├── travel_advisor_agent.py
│   └── final_report_agent.py
│
├── tools/
│   ├── weather_tool.py
│   ├── location_tool.py
│   └── packing_tool.py
│
├── mcp/
│   └── mcp_server.py
│
├── ui/
│   └── ui.py
│
├── .env
├── requirements.txt
└── README.md

⚙️ Installation

Clone Repository

git clone <YOUR_REPOSITORY_URL>
cd travel-planner-mcp

Create Virtual Environment

python -m venv .venv

Activate Environment

Windows:

.venv\Scripts\activate

Linux/macOS:

source .venv/bin/activate

Install Dependencies

pip install -r requirements.txt

🔑 Environment Variables

Create a .env file in the root directory.

GROQ_API_KEY=YOUR_GROQ_API_KEY

▶️ Running the Application

Start FastAPI

uvicorn app:app --reload

Swagger Documentation:

http://127.0.0.1:8000/docs

Start MCP Server

python mcp/mcp_server.py

Start NiceGUI

python ui/ui.py

Application URL:

http://localhost:8080

📸 Example Request

{
  "city": "Ooty",
  "budget": "Medium"
}

📸 Example Response

{
  "weather": {
    "temperature": 18,
    "windspeed": 12
  },
  "packing_list": [
    "Jacket",
    "Water Bottle",
    "Comfortable Shoes"
  ],
  "recommendation": "Good weather for sightseeing and outdoor activities."
}

📚 What I Learned

This project helped me gain hands-on experience with:

• Model Context Protocol (MCP)

• FastMCP Tool Development

• LangGraph State Management

• Agent-Based Workflows

• LLM Tool Calling

• FastAPI Development

• API Integrations

• NiceGUI Dashboard Development

🚀 Future Improvements

• Hotel Recommendation Agent

• Restaurant Recommendation Agent

• Multi-Day Trip Planning

• Budget Estimation

• Google Maps Integration

• Travel Itinerary Generator

• PDF Export

• Multi-Agent Collaboration

👨‍💻 Author

Shyam Sundhar

Computer Science Engineering (AI & ML)

Passionate about:

• Artificial Intelligence

• Machine Learning

• Generative AI

• Agentic AI

• Mobile App Development

• Full Stack Development

🔗 LinkedIn: https://www.linkedin.com/in/shyamgsundhar/

💻 GitHub: https://github.com/shyamgsundhar

⭐ Support

If you found this project useful or interesting, consider giving it a ⭐ on GitHub.

Feedback, suggestions, and contributions are always welcome!