Skip to content

aakar-mutha/agentic-travel-planner

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Virtual Travel Agent

Virtual Travel is a Agentic AI which helps you plan a vacation.

Motivation and Goals

  • I like to travel, however I never know where to go or what to do. I can search the internet for things to do in a particular country but there is always a fear of it not being up to the mark. The itineraries that I find online are not customized to my needs.

  • The goal of this project is to make a team of AI agents that plans a vacation for me given my requirements.

APIs used:

  1. LLM: Anthropic Claud 3 Haiku API.
    • Used to generate queries for tavily, use the information provided by tavily and give a make an itinerary.
  2. Search: Tavily API.
    • Used for getting relevant data for the user's request.

Architecture:

  1. There are 5 different prompts/agents which work together to plan your itinerary.
    • VACATION_PLANNING_SUPERVISOR_PROMPT: Gives the ouline of what the planner has to look for when planning a vacation.
    • PLANNER_ASSISTANT_PROMPT: Uses the haiku model to create prompts that will be useful to form the itinerary.
    • VACATION_PLANNER_PROMPT: Generates the itinerary in the given format using all the information gathered from the earlier steps.
    • PLANNER_CRITIQUE_PROMPT: It gives feedback on the itinerary that is generated. Suggests what else can be considered to plan the vacation.
    • PLANNER_CRITIQUE_ASSISTANT_PROMPT: Calls the tavily api for more information to gather information on suggestions.

Installation

pip install -r requirements.txt

Please add your Tavily and Anthropic API keys as environment variables in the format given in .env.example.

Usage

You have 2 options

  1. Use the python notebook and run it cell by cell.
  2. Use the GUI:
python3 helper.py

Open the link in the browser.

Releases

No releases published

Packages

No packages published