National Park Service MCP Server

An open-source MCP server for managing and processing data related to the National Park Service. This project provides MCP-compliant tools for retriving park information, visitor management, and more.

Getting Started

Installation

Create a virtual environment and activate it:

uv venv .venv source .venv/bin/activate # On Windows use: .venv\Scripts\activate

Sync dependencies from pyproject.toml (if present):

uv sync

To install additional requirements during development:

uv pip install -r requirements.txt

To test the server locally:

```bash

uv run server.py ``` > **Note:** The instructions above are intended for local development and testing. In production, the MCP server is expected to run remotely via Smithery.ai.

MCP Marketplace

This server is open-sourced and available in the MCP Marketplace: View on Smithery.ai

Repository Structure

. ├── src/ │ ├── api_utils.py # To handle API calls to NSP server │ ├── models.py # Pydantic Args to get input from LLMs in specific format ├── server.py # Main python that defines @mcp.tools and its function │── README.md ├── Dockerfile # Docker configuration for containerizing the server ├── smithery.yaml # Smithery configuration ├── LICENSE # MIT license information └── README.md

Environment Variables

Create a .env file in the root directory. Required variables include:

NPS_API_KEY = "***"

Create NPS API key here - https://www.nps.gov/subjects/developer/get-started.htm
NPS API documentation - https://www.nps.gov/subjects/developer/api-documentation.htm