Multi-Agent Travel CLI:

A CLI-based travel itinerary generator using a multi-agent system.

Team Details: Name: DEON GEORGE Institution: St. Joseph’s College of engineering and technology,palai

Challenge Selected: Multi-agent travel itinerary engine

Problem Statement: Kind it difficult to plans the days of the vacation or find difficult in getting the best hotel for your budget and don’t know which is can be the best mode of transportation for thr trip

Solution Overview: The objective of this project is to design and implement a multi-agent travel itinerary planning system that collaboratively generates a complete travel plan based on user inputs such as budget, source location, and travel period. The system is implemented as a command-line interface (CLI) and works offline, without relying on external APIs.

Agentic Design:

This multi-agent has 3 agents Agents

• Destination Agent: Suggests destinations based on budget.
• Transport Agent: Calculates transport and stay costs.
• Itinerary Agent: Generates a day-wise plan. I have done this multi-agent to work in a sequential manner, the orchestrator i.e the main.py connects to the 3 agents. the inputs and the outputs from each agents is collected by the orchestrator and given as the output for the user. Both the destination agent and itinerary agent is LLM based but the transport agent is a rule-based agent which is if the trip is below certain amount it choose train otherwise it choose plane as the mode of transport.

Features: The user enters source city, budget(in INR), number of days the multi agent find out the most favorable destination and the mode of transport along with itinerary plan for the days, in addition it give the hotel cost for stay in that region.

Demo:https://drive.google.com/file/d/1uDhlGLxAQ2_dyMu8QZ9Zhy7D6O3fPTCs/view?usp=sharing

System Architecture:

How to Run the Project:

1. Clone the repository and navigate to the project directory.
2. Install dependencies:

pip install -r requirements.txt

3. Configure Environment:

• Open .env and add your API keys (if applicable in future updates).

Usage

Run the main script:

python main.py

Tech Stack: Python 3 Core language used to implement all agents and system logic Command Line Interface (CLI) Rich (Python Library) Used for: Styled terminal output Panels, tables, and colored text Improved readability and user experience in the terminal.

Limitations & Future Work: Limitations:- The system currently operates only as a Command Line Interface (CLI), which limits accessibility for non-technical users. Travel data such as destinations, transport options, and costs are static and rule-based, not real-time. The system does not support live booking, ticket reservations, or payment integration. Future Work:- The system can be extended into a web application or website using modern web frameworks to improve usability and reach a wider audience. A graphical user interface (GUI) can be developed to replace the CLI, making the system more user-friendly. Real-time travel data APIs can be integrated for live flight, hotel, and pricing information.

Ethics, Safety & Responsible AI: The system provides advisory recommendations only, keeping full decision control with the user. No personal data is stored, shared, or transmitted; all processing is done locally. The application performs no real-world actions such as bookings or payments, ensuring safety. The system uses transparent, rule-based logic, avoiding biased or unpredictable AI behavior. Any future AI integration will follow responsible and ethical usage principles.

Made with ❤️ by Deon George