Skip to content

API_DOCS.md

Latest commit

 

History

History
82 lines (58 loc) · 2.63 KB

API_DOCS.md

File metadata and controls

82 lines (58 loc) · 2.63 KB
Raw

API Documentation for Wayfarer

Overview

Wayfairer provides a modular API through its LangGraph-based workflow. While primarily designed as a Streamlit application, the core functionality can be accessed programmatically through the main workflow engine.

Core Modules

main.py - Workflow Engine

The main workflow orchestrator that manages the multi-agent travel planning process.

Functions:

  • router_agent(state): Classifies user intent as travel-related or chitchat
  • planner_agent(state): Extracts travel details from user input
  • flight_agent(state): Searches for flight options
  • hotel_agent(state): Searches for hotel options
  • itinerary_agent(state): Creates detailed travel itinerary
  • final_agent(state): Formats final response for user
  • chitchat_agent(state): Handles non-travel conversations

State Structure (TravelState):

{
  "messages": List[AnyMessage],  // Conversation history
  "user_query": str,             // Original user input
  "destination": str,            // Target destination
  "departure_city": str,         // Starting location
  "dep_iata": str,               // Departure airport code
  "arr_iata": str,               // Arrival airport code
  "flight_search_query": str,    // Flight search parameters
  "hotel_search_query": str,     // Hotel search parameters
  "flight_results": str,         // Raw flight data
  "hotel_results": str,          // Raw hotel data
  "itinerary": str,              // Generated itinerary
  "llm_calls": int,              // Counter for LLM usage
  "is_travel_related": bool      // Intent classification result
}

app.py - Streamlit Interface

Handles the user interface and real-time visualization of the agent workflow.

Key Components:

  • Hero section with animated gradient background
  • Destination strip with popular locations
  • Input area with quick suggestion chips
  • Real-time agent tracking visualization
  • Metrics display (agents run, LLM calls, status)
  • Boarding pass style results presentation
  • Download functionality (Markdown/PDF)

Tools

flight_tool.py - Flight Search Functionality

Provides flight search capabilities with fallback mechanisms.

Functions:
  • search_flights(query: str, dep_iata: str = None, arr_iata: str = None) -> str
    • Searches for flights using AviationStack API
    • Falls back to Tavily search when API unavailable
    • Returns formatted flight information string

tavily_tool.py - Tavily Search Wrapper

Provides web search capabilities for hotel and travel information.

Functions:
  • tavily_search(query: str) -> str

<tool_call> <function=Bash> <parameter=command> git add API_DOCS.md