AI Travel Agent

An intelligent, multi-agent system for planning your travel, built with the @iqai/adk library.

Flights • Hotels • Smart Assistance

---

Welcome! This project is an AI-powered travel assistant designed to make trip planning a breeze. It uses a sophisticated, hierarchical agent system to handle different aspects of your journey. You can ask it to find flights, search for hotels, and it will intelligently figure out the details for you.

• Hierarchical Agent System: A main root_travel_agent orchestrates tasks, delegating to specialized agents.
• ✈️ Flight Agent: Finds the best flight options for your trip.
  • Searches for available flights between cities.
  • Analyzes flight prices to find good deals.
• 🏨 Hotel Agent: Helps you find the perfect place to stay.
  • Searches for hotels at your destination.
  • Provides detailed information about specific hotels.
• 📍 Smart Location Handling: Automatically converts city names (like "Paris") into IATA codes (like "CDG") needed for flight and hotel searches.

Ready to try it out? Here’s how to get the agent running on your local machine.

git clone https://github.com/YOUR_USERNAME/iq-flights.git
cd iq-flights

This project uses pnpm for fast and efficient package management. This will install all the necessary packages listed in package.json.

pnpm install

You can also use this template directly by copying the files, but using the CLI is recommended for best results.

Default (Production/Development) Route

To run your agent in production or for standard development, use:

pnpm dev

Fast Iteration & Agent Setup (ADK CLI)

For rapid prototyping, interactive testing, or initial agent setup, use the ADK CLI:

adk run   # Interactive CLI chat with your agents
adk web   # Web interface for easy testing and demonstration

The main agent code lives in index.ts where the subagents live inside the agents folder. The agents/agent.ts file is compatible with the ADK CLI for easy testing.

├── src/
│   ├── agents/
│   │   ├── agent.ts          # Root agent (ADK CLI compatible)
│   │   ├── joke-agent/       # Joke-telling sub-agent
│   │   │   ├── agent.ts
│   │   │   └── tools.ts
│   │   └── weather-agent/    # Weather information sub-agent
│   │       ├── agent.ts
│   │       └── tools.ts
│   ├── env.ts                # Environment variable validation
│   └── index.ts              # Main execution entry point

Make sure to configure your environment variables:

cp .env.example .env

This starter includes:

• GitHub Actions: CI/CD pipeline
• 📦 PNPM: Fast package manager
• 🤖 ADK CLI: Interactive testing with adk run and adk web

Traditional Testing: Run pnpm dev to execute the sample questions.

Interactive Testing with ADK CLI:

1. Install: npm install -g @iqai/adk-cli
2. Run: adk run for CLI chat or adk web for web interface
3. Perfect for development, testing, and demonstrating your agent's capabilities

1. Create new agents in the src/agents/ directory
2. Add tools to your agents in the tools/ subdirectory
3. Configure services in the src/services/ directory
4. Update environment variables in src/env.ts

• ADK Documentation
• ADK GitHub Repository

If you encounter any issues or have questions:

• 📝 Create an issue
• 💬 Start a discussion