home / mcp / mcp travel server
A mcp for your Amazon Rain forest Adventure!
Configuration
View docs{
"mcpServers": {
"ajey95-mcp": {
"command": "uv",
"args": [
"run",
"mcp",
"install",
"file.py"
],
"env": {
"ORS_API_KEY": "YOUR_OPENROUTESERVICE_API_KEY",
"GEOAPIFY_API_KEY": "YOUR_GEOAPIFY_API_KEY",
"OPENTRIPMAP_API_KEY": "YOUR_OPENTRIPMAP_API_KEY"
}
}
}
}You can run an MCP server that connects Claude to a travel assistant experience. This server enables Claude to fetch travel routes, nearby hotels, adventure spots, cultural sites, and essential precautions by using configured data sources and endpoints. It provides a structured set of tools that Claude can call to guide users through planning trips and exploring destinations.
You will connect Claude to your MCP server as a tool provider. Start your server, then add its endpoint to Claude so the assistant can discover and use its tools. The most common workflow is to run the server locally during development and point Claude at the local endpoint.
# Prerequisites
- Python 3.9+ (recommended)
- Git
- Optional: Python virtual environment tool (venv)
# Clone the MCP server repository
git clone https://github.com/Ajey95/MCP.git
cd MCP
# On Windows, install the UV tool (server runner)
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
# If you prefer other shells, follow the platform-specific guidance from your environment.
# Optional: create and activate a virtual environment
python -m venv venv
# On Windows
venv\Scripts\activate
# On Unix
source venv/bin/activate
# Add your API keys in a .env file
# Example entries
ORS_API_KEY=your_openrouteservice_api_key
GEOAPIFY_API_KEY=your_geoapify_api_key
OPENTRIPMAP_API_KEY=your_opentripmap_api_key
# Install MCP CLI tooling (if not already installed)
install mcp
install mcp[cli]
# Final project dependencies
pip install -r requirements.txt
# Run the MCP server (final start command from the flow)
# This starts the runtime that Claude will connect to
uv run mcp install file.py
# Continue building and refining your server as needed
```
Note: The start command shown above runs the MCP server locally and exposes the plugin endpoint for Claude to discover and register its tools.Configuration and security: - Ensure your API keys are stored securely in a .env file and not committed to source control. - Limit access to the locally exposed endpoint during development. Consider using a VPN or host firewall rules for exposure in production. Usage notes: - Start the local MCP server with the provided start command, then use Claude to add the plugin endpoint and install tools such as get_travel_route, get_nearby_hotels, get_adventure_spots, get_monuments_and_food, get_precautions, and greet_user. - When you configure Claude, point it to the local endpoint shown in your terminal (for example, http://localhost:3333/.well-known/ai-plugin.json). Troubleshooting: - If Claude cannot discover tools, verify the server is running and the plugin endpoint is reachable. - Check that your .env values are loaded by the runtime; log any missing API keys and re-run the server. - Review dependency installation messages for any missing packages and install them as needed. Notes: - The server exposes a set of tools that map to travel-related actions, enabling Claude to respond with routes, hotel suggestions, adventure spots, and cultural/site recommendations.
This MCP server provides a set of tools that Claude can call: - get_travel_route: Compute travel directions and travel time between two locations. - get_nearby_hotels: Discover nearby hotels using the configured data sources. - get_adventure_spots: Find hikes, nature trails, and adventure spots near a location. - get_monuments_and_food: Explore food options, monuments, and cultural sites near a locale. - get_precautions: Retrieve essential travel precautions before visiting the rainforest. - greet_user: Provide a friendly greeting and introduce available capabilities.
To connect Claude, run your MCP server locally and then add its endpoint in Claude. Use the CLI-provided endpoint shown in the terminal (for example, http://localhost:3333/.well-known/ai-plugin.json). Claude will automatically discover and install the available tools, including get_travel_route, get_nearby_hotels, get_adventure_spots, get_monuments_and_food, get_precautions, and greet_user.
Compute travel directions and duration between two locations using the configured routing data sources.
Discover nearby hotels around a location using Geoapify and related hotel data sources.
Find hikes, nature trails, and adventure spots near a given area.
Explore cultural sites, monuments, and nearby food options.
Provide essential travel precautions for visiting the rainforest.
Deliver a friendly greeting and overview of available tools.