This project provides an AI Travel Agent built with LangGraph, designed to assist users with travel planning tasks like finding flights and hotels, and automating email communications. It targets users seeking a practical, stateful, and interactive AI assistant for travel arrangements, offering a seamless experience with human oversight.

How It Works

The agent utilizes LangGraph to orchestrate a multi-LLM workflow, enabling stateful interactions and dynamic LLM selection for different tasks. It incorporates a human-in-the-loop mechanism for critical actions, such as reviewing travel plans before sending emails, and leverages tools for data retrieval and communication.

Quick Start & Requirements

• Install dependencies: poetry install --sync
• Set API keys in a .env file (OpenAI, SerpApi, SendGrid, LangChain).
• Run the chatbot: streamlit run app.py
• Requires Python 3.11.9 (via pyenv).
• Documentation: Medium Article

Highlighted Details

• Stateful interactions remember user context across conversations.
• Human-in-the-loop control for user review of critical actions.
• Dynamic LLM switching for task-specific optimization.
• Email automation for sending personalized travel plans.

Maintenance & Community

No specific community channels or notable contributors are mentioned in the README.

Licensing & Compatibility

Distributed under the MIT License, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

The project relies on external APIs (Google Flights, Google Hotels, OpenAI, SerpApi, SendGrid), requiring valid API keys and potentially incurring costs. The data fetching is dependent on these third-party services.