🌿 Amazon Rainforest Travel Guide (Claude Agent + MCP)
An AI-powered travel assistant connected to Claude via MCP, designed to help users explore the Amazon Rainforest — from planning routes to discovering nearby hotels and adventure spots.
🚀 Features
- 🌍 Get directions from any location to Manaus, Brazil
- 🏨 Discover nearby hotels using Geoapify
- 🥾 Find adventure spots like hikes and nature trails
- 🏛️ Explore food, monuments, and cultural sites
- ⚠️ Get essential travel precautions
- 🤖 Powered by Claude + FastMCP
🛠️ Tech Stack
- Claude (Anthropic Desktop Client)
- FastMCP for Claude tool registration
- Python + HTTPX + dotenv
- OpenTripMap, Geoapify, OpenRouteService APIs
📦 Installation
- Clone the repo:
git clone https://github.com/Ajey95/MCP.git
cd MCP
- commands:
# On Windows.
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
uv init projectname
Create a virtual environment (optional but recommended):
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
Add your API keys in a .env file:
ORS_API_KEY=your_openrouteservice_api_key
GEOAPIFY_API_KEY=your_geoapify_api_key
OPENTRIPMAP_API_KEY=your_opentripmap_api_key
- mcp command:
install mcp
install mcp[cli]
- Final:
pip install -r requirements.txt
To run uv run mcp install file.py
#Continue building!
🧠 Connecting Claude to Your MCP Server
Run your server:
uv run mcp install file.py
Open the Claude desktop app
Click on Add Tool
Paste your MCP endpoint (shown in terminal, e.g., http://localhost:3333/.well-known/ai-plugin.json)
Claude will scan and install tools like:
- get_travel_route
- get_nearby_hotels
- get_adventure_spots
- get_monuments_and_food
- get_precautions
- greet_user
🗣️ Sample Prompts to Try in Claude
- "What's the distance and travel time from Rio de Janeiro to the Amazon Rainforest?"
- "Find top hotels near Manaus."
- "Suggest some adventure spots near Manaus."
- "What cultural sites can I visit in the Amazon?"
- "Give me precautions before visiting the rainforest."
🤝 Contributing
Pull requests welcome. For major changes, please open an issue first.
📄 License
MIT