NPS Explorer MCP Server
Provides LLM context on U.S. national parks by integrating National Park Service, Recreation.gov, and weather APIs, enabling queries about park info, trails, alerts, events, weather forecasts, and more via tool calls.
README
NPS Explorer MCP Server 🔭
Overview
This repository is a modified version of Cloudflare’s remote-MCP-GitHub-OAuth demo, extended into a National Parks Service Explorer. It provides an LLM context on U.S. national parks by integrating:
- The National Park Service API
- The Recreation.gov API
- A Weather API (I chose https://www.weatherapi.com/)
It runs as a Model Context Protocol (MCP) server on Cloudflare Workers, letting you query park overviews, trails, alerts, events, weather forecasts, and more via simple tool calls to the LLM.
Looking for the original demo?
See Cloudflare’s instructions here.Want to add this tool directly to your Claude Desktop?
Checkout the instructions hereWant to clone their example directly? Use their command line to get a server going on your machine:
npm create cloudflare@latest -- my-mcp-server --template=cloudflare/ai/demos/remote-mcp-github-oauth
📋 Prerequisites
- A Cloudflare account with Workers and KV enabled
- Node.js v18.20.4+ and npm
- Your own API keys for:
- National Park Service API
- Recreation.gov API
- Weather API (weatherapi)
⚙️ Setup
-
Clone this repo
git clone https://github.com/Kyle-Ski/nps-explorer-mcp-server.git cd nps-explorer-mcp-server -
Copy and fill
.dev.varscp .dev.vars.example .dev.varsThen open .dev.vars and populate with your own secrets.
- Install dependencies
npm i
- Run Locally
npm start
🚀 Deployment
The deploy command is using Wrangler to deploy your server to Cloudflare Workers! Cloudflare will also use their Durable Objects and Workers KV.
npm run deploy
🧰 Tools
| Tool Name | Description | Status |
|---|---|---|
getParkInfo |
Get comprehensive information about a national park including both static details and current conditions | working |
getTrailInfo |
Get detailed information about trails (difficulty, length, elevation gain, current conditions) | 🚧 under construction |
findParks |
Find national parks based on criteria such as state, activities, or amenities | working |
getParkAlerts |
Get current alerts, closures, and notifications for specified parks | working |
getParkEvents |
Get upcoming events at parks including ranger talks, guided hikes, and educational programs | working |
findNearbyRecreation |
Find recreation areas and camping options near a given location | ⚠️ not working correctly |
planParkVisit |
Get recommendations for the best time to visit a park based on historical and forecast weather | working |
getParkWeatherForecast |
Get detailed weather forecast for a national park by park code | working |
getCampgrounds |
List campgrounds within a given national park with detailed information | working |
🧪 Testing
Once your server is up and running, use the MCP Server Inspector Tool to make sure your server can connect and show you the tools and resources it has access to.
npx @modelcontextprotocol/inspector
🛜 For Production
All of this is in the original Cloudflare README.md
Cloudflare recommends creating a local and remote GitHub OAuth App:
- For the Homepage URL, specify
https://mcp-github-oauth.<your-subdomain>.workers.dev - For the Authorization callback URL, specify
https://mcp-github-oauth.<your-subdomain>.workers.dev/callback - Note your Client ID and generate a Client secret and add them via Wrangler.
wrangler secret put GITHUB_CLIENT_ID
wrangler secret put GITHUB_CLIENT_SECRET
wrangler secret put COOKIE_ENCRYPTION_KEY # add any random string here e.g. openssl rand -hex 32
Set up a KV namespace
- Create the KV namespace: wrangler kv:namespace create "OAUTH_KV"
- Update the Wrangler file with the KV ID
📚 Reference
- Original Cloudflare Demo README: https://github.com/cloudflare/ai/tree/main/demos/remote-mcp-github-oauth#readme
- Model Context Protocol: https://modelcontextprotocol.io/introduction
- Connecting an agent to your server:
🧑🔧 Troubleshooting and Common Issues
Check out the MCP troubleshooting docs
Connecting to Claude Desktop Fails:
- Most commonly due to Node incompatibility issues, use node v18.20.4+ locally and checkout this guide I wrote.
🤝 Contributing
-
Fork this repository
-
Create a feature branch (git checkout -b feature-name)
-
Commit your changes (git commit -m 'Add new tool')
-
Push to the branch (git push origin feature-name)
-
Open a Pull Request
⚖️ License
his project is licensed under the MIT License. See LICENSE for details.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。