Travel-Seek

Travel-Seek is an AI-powered travel planning assistant that helps you create personalized travel itineraries and answer travel-related questions. The application leverages a coordinated multi-agent team (powered by Agno + Groq) and optionally enriches results with real Google Places & Directions data via a custom MCP server plus verified OpenWeather forecasts.

Features

• Multi-Agent Travel Planning: Four specialized AI agents (Research, Itinerary, Budget, Local Expert) coordinate to produce comprehensive travel plans.
• Google Places Integration: Real ratings, reviews, hours, and directions via a custom MCP server backed by the Google Maps API.
• OpenWeather Integration: Verified current conditions and 5-day forecasts, with explicit long-range limitations instead of fabricated weather detail.
• Conversational Planning: Chat naturally to build and refine your plan with session memory.
• Graceful Fallback: Works without a Google Maps key — falls back to DuckDuckGo search with a visible warning.
• Standalone MCP Server: The Places MCP server works independently with Claude Desktop or any MCP-compatible client.

Installation

1. Clone the Repository

git clone https://github.com/Mukku27/Travel-Seek.git
cd Travel-Seek

2. Create and Activate a Virtual Environment

python3 -m venv .venv
source .venv/bin/activate  # On Windows use `.venv\Scripts\activate`

3. Install Dependencies

pip install -r requirements.txt

4. Set Up Environment Variables

Create a .env file in the root directory:

GROQ_API_KEY='your_groq_api_key'
GOOGLE_MAPS_API_KEY='your_google_maps_api_key'   # Optional — enables Places & Directions
OPENWEATHERMAP_API_KEY='your_openweather_api_key' # Optional — enables verified weather tools
SERPI_API_KEY='your_serpi_api_key'

# Optional: enable Tavily search alongside SerpAPI/DuckDuckGo
USE_TAVILY='true'
TAVILY_API_KEY='your_tavily_api_key'

The GOOGLE_MAPS_API_KEY requires the following APIs enabled in your Google Cloud project:

• Places API
• Directions API
• Geocoding API

If the key is not set, the app works normally using DuckDuckGo for all research. If USE_TAVILY is enabled and TAVILY_API_KEY is present, the research agent can use Tavily alongside DuckDuckGo. If OPENWEATHERMAP_API_KEY is set, the planner can use real OpenWeather current conditions and 5-day forecasts. For trips beyond that forecast window, the app tells the traveler to recheck later instead of inventing day-level weather.

Usage

Run the Streamlit Application

streamlit run app.py

Open your browser at http://localhost:8501, enter your trip details in the sidebar, and either generate a one-shot plan or chat interactively.

Run Tests

pytest tests/ -v

Project Structure

Travel-Seek/
├── agents/
│   ├── __init__.py
│   ├── travel_team.py        # Team orchestrator with MCP lifecycle
│   ├── research_agent.py     # DuckDuckGo + optional MCP tools
│   ├── itinerary_agent.py    # Day-by-day planner + optional Directions
│   ├── budget_agent.py       # Cost estimates via DuckDuckGo
│   └── local_expert_agent.py # Cultural tips via DuckDuckGo
├── tools/
│   ├── __init__.py
│   └── weather_tool.py       # OpenWeather toolkit for live weather guidance
├── mcp_servers/
│   ├── __init__.py
│   ├── config.py             # API key management & client factory
│   ├── places_tools.py       # Core Google Maps tool logic
│   └── places_server.py      # FastMCP server (3 tools)
├── tests/
│   ├── test_places_tools.py
│   ├── test_places_server.py
│   ├── test_mcp_config.py
│   ├── test_travel_team_fallback.py
│   ├── test_weather_tool.py
│   ├── test_weather_tool_live.py
│   ├── test_models.py
│   └── test_utils.py
├── app.py                    # Streamlit entry point
├── config.py                 # App configuration
├── models.py                 # Pydantic models
├── prompt.py                 # Prompt templates
├── utils.py                  # Response cleaning utilities
├── requirements.txt
└── README.md

MCP Server

The mcp_servers/ package exposes three Google Maps tools as a standalone MCP server:

Tool	Description
search_places	Geocodes a location, then runs a Places Nearby Search. Returns top 5 results with name, rating, address, price level, etc.
get_place_details	Fetches rich details for a place: reviews, hours, phone, website, geometry.
get_directions	Computes directions between two locations with distance, duration, and step-by-step instructions.

Run Standalone (stdio)

python -m mcp_servers.places_server

Run with SSE Transport

python -m mcp_servers.places_server --transport sse --port 8000

Claude Desktop Configuration

Add this to your Claude Desktop MCP config:

{
  "mcpServers": {
    "travel-seek-places": {
      "command": "python",
      "args": ["-m", "mcp_servers.places_server"],
      "env": {
        "GOOGLE_MAPS_API_KEY": "your_key"
      }
    }
  }
}

License

This project is licensed under the MIT License. See the LICENSE file for details.

Contributing

Contributions are welcome! Please open an issue or submit a pull request for any improvements or bug fixes.