AI Trip Planner

Autonomous AI travel planning with a full web interface, powered by CrewAI and GPT-4

AI Trip Planner orchestrates three specialized AI agents to research destinations, compile local insider guides, and generate complete day-by-day travel itineraries. Enter your travel preferences and receive a fully personalized plan with hotel picks, restaurant recommendations, a packing list, and a detailed budget breakdown.

Screenshots

| Landing Page | Form Filled | |:---:|:---:| | | |

| Results View | Day-by-Day Itinerary | |:---:|:---:| | | |

Features

| Feature | Description | |---------|-------------| | Multi-agent planning | City Selector, Local Expert, and Travel Concierge collaborate sequentially | | Real-time web search | Agents query live travel data, prices, and events via Serper API | | Budget-aware itineraries | Tailored plans for Budget, Mid-range, and Luxury travelers | | Group travel support | Adjusts costs and recommendations for any number of travelers | | Streamlit web UI | Clean browser-based interface with sidebar API configuration | | CLI mode | Full command-line interface for terminal users | | Downloadable plans | Export your itinerary as a Markdown file |

Architecture

flowchart TD
    A[User Input] --> B[City Selection Agent]
    B -->|Serper Search| C[Destination Analysis Report]
    C --> D[Local Expert Agent]
    D -->|Serper Search| E[City Insider Guide]
    E --> F[Travel Concierge Agent]
    F -->|Serper Search| G[Live Research]
    F -->|Calculator| H[Budget Math]
    G & H --> I[Complete Day-by-Day Itinerary]

Project Structure

.
├── src/
│   ├── agents.py          # Three specialized CrewAI agents
│   ├── tasks.py           # Task definitions with expected outputs
│   ├── crew.py            # TripCrew orchestrator
│   └── tools/
│       ├── search.py      # Serper web search tool
│       └── calculator.py  # Safe arithmetic evaluation tool
├── app.py                 # Streamlit web interface
├── cli.py                 # Command-line interface
├── .env.example           # Environment variable template
└── requirements.txt

Overview

This project comprises several Python modules:

• crewai: Core framework for managing the trip planning agent crew with sequential task execution.
• src/agents: Specialized agents for city selection, local expertise, and travel concierge services.
• src/tasks: Task definitions for destination analysis, city research, and itinerary creation.
• langchain: Integration with language models via langchain-openai, supporting OpenAI GPT models.

Getting Started

Prerequisites

• Python 3.10 or higher
• OpenAI API key
• Serper API key (free tier available, 2 500 free queries/month)

Installation

git clone https://github.com/punyamodi/AI-Trip-Planner-using-CrewAI.git
cd AI-Trip-Planner-using-CrewAI

python -m venv venv
# Windows
venv\Scripts\activate
# macOS / Linux
source venv/bin/activate

pip install -r requirements.txt

Configuration

Copy the example environment file and add your API keys:

cp .env.example .env

Edit .env:

OPENAI_API_KEY=your_openai_api_key
SERPER_API_KEY=your_serper_api_key
OPENAI_MODEL=gpt-4o-mini

API keys can also be entered directly in the Streamlit sidebar without creating a .env file.

Usage

Web Interface

streamlit run app.py

Open http://localhost:8501 in your browser. Fill in your trip details and click Plan My Trip.

Command Line

python cli.py

Follow the prompts. The completed itinerary is printed to the terminal and saved as trip_itinerary.md.

Programmatic

from src.crew import TripCrew

crew = TripCrew(
    origin="New York",
    destinations="Tokyo, Bangkok, Bali",
    date_range="September 5 to September 20, 2025",
    interests="street food, temples, photography, hiking",
    budget="Mid-range",
    travelers=2,
)

result = crew.run()
print(result)

Agent Workflow

| Agent | Responsibility | Tools | |-------|---------------|-------| | City Selection Expert | Picks the best destination by comparing weather, events, and flight costs | Web Search | | Local Expert | Compiles a detailed insider guide: attractions, neighborhoods, food, culture | Web Search | | Travel Concierge | Builds the full itinerary with hotels, restaurants, budget, and packing list | Web Search, Calculator |

Example Input

| Field | Example Value | |-------|--------------| | Traveling from | New York | | Destinations | Tokyo, Bangkok, Bali | | Travel dates | September 5 to September 20, 2025 | | Interests | Street food, temples, photography, nature | | Budget | Mid-range | | Travelers | 2 |

Configuration Reference

| Variable | Description | Default | |----------|-------------|---------| | OPENAI_API_KEY | OpenAI API key | Required | | SERPER_API_KEY | Serper search API key | Required | | OPENAI_MODEL | OpenAI model name | gpt-4o-mini |

Contributing

Contributions are welcome. Open an issue or submit a pull request on GitHub.

License

MIT License