avalanche-org-mcp-server
Enables LLMs to query current and historic avalanche danger ratings by location and retrieve raw forecast GeoJSON from the Avalanche.org public API.
README
avalanche-org-mcp-server
A minimal Model Context Protocol (MCP) server that wraps the Avalanche.org Public API map-layer endpoints. It lets LLMs look up avalanche danger ratings by location, retrieve raw forecast GeoJSON, and query historic conditions.
Features
- Danger lookup by lat/lon — find the avalanche zone for any point and get its current danger rating
- Historic danger lookup — same as above, but for a specific past date
- Raw map-layer GeoJSON — full FeatureCollection for all avalanche centers, or scoped to one center
Data sourced from the Avalanche.org Public API. See the API docs for details.
Quick Start
Claude Desktop (npx — recommended)
Add this to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"avalanche-org": {
"command": "npx",
"args": ["-y", "avalanche-org-mcp-server"]
}
}
}
No install required. Claude will download and run it automatically.
Global npm install
npm install -g avalanche-org-mcp-server
{
"mcpServers": {
"avalanche-org": {
"command": "avalanche-org-mcp-server"
}
}
}
Docker
docker pull ghcr.io/haydenwade/avalanche-org-mcp-server:latest
{
"mcpServers": {
"avalanche-org": {
"command": "docker",
"args": ["run", "--rm", "-i", "ghcr.io/haydenwade/avalanche-org-mcp-server:latest"]
}
}
}
From source
git clone https://github.com/haydenwade/avalanche-org-mcp-server.git
cd avalanche-org-mcp-server
npm install
npm run build
{
"mcpServers": {
"avalanche-org": {
"command": "node",
"args": ["/absolute/path/to/avalanche-org-mcp-server/dist/src/index.js"]
}
}
}
Tools
avalanche_danger_rating_by_point
Get the current avalanche danger rating for a lat/lon point. Returns the zone the point falls in, or the nearest zone if preferNearest is true.
| Parameter | Type | Required | Description |
|---|---|---|---|
lat |
number | yes | Latitude in decimal degrees |
lon |
number | yes | Longitude in decimal degrees |
preferNearest |
boolean | no | Fall back to nearest zone if point is outside all polygons (default: true) |
centerId |
string | no | Scope search to a specific avalanche center (e.g. "UAC", "CBAC") |
Example input:
{ "lat": 40.5763, "lon": -111.7522, "preferNearest": true, "centerId": "UAC" }
<details> <summary><strong>Example output</strong></summary>
{
"match": "inside_zone",
"distance_km": 0,
"distance_miles": 0,
"zone": {
"id": "zone-1",
"name": "Salt Lake",
"state": "UT"
},
"center": {
"id": "UAC",
"name": "Utah Avalanche Center",
"timezone": "America/Denver",
"link": "https://utahavalanchecenter.org"
},
"danger": {
"level": 3,
"label": "Considerable",
"color": "#f1a302"
},
"travel_advice": "Dangerous avalanche conditions. Careful snowpack evaluation, cautious route-finding and conservative decision-making essential.",
"forecast_url": "https://utahavalanchecenter.org/forecast/salt-lake",
"validity": {
"start_date": "2026-03-22",
"end_date": "2026-03-23"
},
"warning": null
}
</details>
historic_avalanche_danger_rating_by_point
Same as above, but for a specific historic date. Returns all the same fields plus day.
| Parameter | Type | Required | Description |
|---|---|---|---|
lat |
number | yes | Latitude in decimal degrees |
lon |
number | yes | Longitude in decimal degrees |
day |
string | yes | Date in YYYY-MM-DD format |
preferNearest |
boolean | no | Fall back to nearest zone (default: true) |
centerId |
string | no | Scope to an avalanche center |
Example input:
{ "lat": 40.5763, "lon": -111.7522, "day": "2025-02-24", "preferNearest": true }
<details> <summary><strong>Example output</strong></summary>
{
"match": "inside_zone",
"distance_km": 0,
"distance_miles": 0,
"zone": { "id": "zone-1", "name": "Salt Lake", "state": "UT" },
"center": { "id": "UAC", "name": "Utah Avalanche Center", "..." : "..." },
"danger": { "level": 3, "label": "Considerable", "color": "#f1a302" },
"travel_advice": "Dangerous avalanche conditions. ...",
"forecast_url": "https://utahavalanchecenter.org/forecast/salt-lake",
"validity": { "start_date": "2025-02-24", "end_date": "2025-02-25" },
"warning": null,
"day": "2025-02-24"
}
</details>
raw_map_layer
Returns the raw Avalanche.org map-layer GeoJSON FeatureCollection for all avalanche centers.
| Parameter | Type | Required | Description |
|---|---|---|---|
day |
string | no | Historic date in YYYY-MM-DD format |
<details> <summary><strong>Example output</strong></summary>
{
"geojson": {
"type": "FeatureCollection",
"features": [ "... full GeoJSON features ..." ]
}
}
</details>
raw_map_layer_by_avalanche_center
Returns the raw map-layer GeoJSON FeatureCollection for a single avalanche center.
| Parameter | Type | Required | Description |
|---|---|---|---|
centerId |
string | yes | Avalanche center ID (e.g. "CBAC", "NWAC", "UAC") |
day |
string | no | Historic date in YYYY-MM-DD format |
<details> <summary><strong>Example output</strong></summary>
{
"geojson": {
"type": "FeatureCollection",
"features": [ "... full GeoJSON features ..." ]
}
}
</details>
Development
Requires Node.js >= 18.
npm install
npm run build
npm test
To interactively test tools with the MCP Inspector:
npx @modelcontextprotocol/inspector node dist/src/index.js
Project structure
src/
index.ts # Entry point — server setup + stdio transport
tools.ts # Tool registration
constants.ts # API URLs and timeouts
types.ts # GeoJSON type definitions
api/
mapLayer.ts # Map-layer fetch + GeoJSON normalization
lib/
geometry.ts # Point-in-polygon, haversine distance, bounds
dangerLookup.ts # Danger rating lookup logic
validation.ts # Date and coordinate validation
test/
*.test.ts # Tests (node:test)
Contributing
- Fork the repo
- Create a feature branch (
git checkout -b my-feature) - Make your changes and add tests
- Run
npm testto make sure everything passes - Open a pull request
License
MIT
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。