geosphere-mcp-server
MCP server for worldwide weather data, using high-resolution GeoSphere Austria data for the Alpine region and Open-Meteo elsewhere. Provides current conditions, hourly forecasts, and daily outlooks with compact emoji-markdown output.
README
geosphere-mcp-server
<!-- mcp-name: io.github.slettmayer/geosphere-mcp-server -->
MCP server for weather: current conditions, hourly forecasts, and multi-day outlooks for any location worldwide, via the Model Context Protocol.
In Austria and the Alpine region it serves high-resolution GeoSphere Austria data — the AROME numerical forecast, the INCA analysis/nowcast, and the C-LAEF ensemble for precipitation probability. Everywhere else it falls back automatically to Open-Meteo, so it is a drop-in worldwide weather source. Every response states which source produced it.
Output is compact emoji-markdown with metric units — built for smart-home and voice-assistant LLM pipelines where a terse, readable answer beats a JSON blob. Weather conditions are derived from physical parameters and reported with the Home Assistant condition vocabulary (sunny, partlycloudy, rainy, snowy, …).
Coverage
| Where | get_current_weather |
get_hourly_forecast |
get_daily_forecast |
|---|---|---|---|
| Austria | GeoSphere INCA + nowcast + AROME | GeoSphere AROME (≤60 h) + C-LAEF probability | Open-Meteo (1–16 days) |
| Alps (non-AT) | GeoSphere AROME only | GeoSphere AROME (≤60 h) + C-LAEF probability | Open-Meteo (1–16 days) |
| Rest of world | Open-Meteo | Open-Meteo (≤48 h) | Open-Meteo (1–16 days) |
Coverage is detected automatically: the server tries GeoSphere first and falls back to Open-Meteo when the point is outside the AROME grid — no bounding box to configure. The daily forecast always uses Open-Meteo (GeoSphere publishes no forecasts beyond ~60 h).
Installation
Pass decimal latitude/longitude to every tool. There is no geocoder in the server — the calling LLM geocodes city names to coordinates itself.
Claude Desktop
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"geosphere": {
"command": "uvx",
"args": ["geosphere-mcp-server"]
}
}
}
Claude Code
claude mcp add geosphere -- uvx geosphere-mcp-server
From source (development)
{
"mcpServers": {
"geosphere": {
"command": "uvx",
"args": ["--from", "/path/to/geosphere-mcp-server", "geosphere-mcp-server"]
}
}
}
Home Assistant
It is a standard stdio MCP server, so it runs anywhere an stdio MCP server can be hosted — including alongside Home Assistant's Assist pipeline (register it the same way as any other stdio MCP server). Give the voice agent the coordinates of the places it should answer for, or let it geocode names.
Tools
get_current_weather
Current conditions for a point. GeoSphere (INCA/AROME) inside coverage, Open-Meteo elsewhere.
| Parameter | Type | Default | Description |
|---|---|---|---|
latitude |
float | required | Decimal latitude (e.g. 48.2208) |
longitude |
float | required | Decimal longitude (e.g. 16.3738) |
# Current Weather at 48.2208, 16.3738
🌡️ Temperature: 24.7°C (feels like 24.2°C)
🌤️ Condition: partlycloudy
💧 Humidity: 52%
💨 Wind: 2.9 m/s from 135° (gusts 7 m/s)
🌧️ Precipitation (last hour): 0 mm
📊 Pressure: 1016 hPa
☁️ Cloud cover: 30%
🕐 Timezone: Europe/Vienna (CEST)
📡 Source: GeoSphere (INCA + nowcast + AROME) — observed 15:20
Outside GeoSphere coverage the same tool answers from Open-Meteo (sunrise/sunset and the point's own timezone included):
# Current Weather at 38.7223, -9.1393
🌡️ Temperature: 26.1°C (feels like 27.0°C)
🌤️ Condition: cloudy
💧 Humidity: 58%
💨 Wind: 4.5 m/s from 315° (gusts 9 m/s)
📊 Pressure: 1014 hPa
☁️ Cloud cover: 90%
🌅 Sunrise: 06:24
🌇 Sunset: 20:52
🕐 Timezone: Europe/Lisbon (WEST)
📡 Source: Open-Meteo — observed 14:00
get_hourly_forecast
Hour-by-hour forecast. GeoSphere AROME (with C-LAEF precipitation probability) up to ~60 h inside coverage; Open-Meteo up to 48 h elsewhere.
| Parameter | Type | Default | Description |
|---|---|---|---|
latitude |
float | required | Decimal latitude (e.g. 48.2208) |
longitude |
float | required | Decimal longitude (e.g. 16.3738) |
hours |
int | 24 | Forecast hours (clamped to 1–60 on GeoSphere, 1–48 on the fallback) |
start |
string | now | Optional ISO 8601 start (e.g. 2026-07-22T15:00); forecast begins at/after this instant |
# 3-Hour Forecast for 48.2208, 16.3738
AROME model, reference 2026-07-22 12:00 CEST · Source: GeoSphere (AROME + C-LAEF ensemble)
Wed 2026-07-22
15:00: 24.7°C — partlycloudy, wind 3 m/s
16:00: 23.9°C — rainy, 1.2 mm (70% chance), wind 4 m/s
17:00: 22.5°C — cloudy, wind 3 m/s
The hours are grouped under a day-divider header (%a %Y-%m-%d in the point's local timezone) that repeats whenever the local date changes, so a window crossing midnight stays unambiguous. Requesting more hours than the AROME horizon provides appends a note suggesting get_daily_forecast for days further ahead. Dry hours omit the precipitation and probability parts.
get_daily_forecast
Multi-day outlook, always from Open-Meteo (worldwide, including Austria).
| Parameter | Type | Default | Description |
|---|---|---|---|
latitude |
float | required | Decimal latitude (e.g. 48.2208) |
longitude |
float | required | Decimal longitude (e.g. 16.3738) |
days |
int | 7 | Forecast days from today (clamped to 1–16). Ignored when start_date/end_date are given |
start_date |
str | — | Optional first forecast day, ISO YYYY-MM-DD |
end_date |
str | — | Optional last forecast day (inclusive), ISO YYYY-MM-DD; defaults to start_date. Span capped at 16 days |
Request either a count of days from today (days) or an explicit calendar range (start_date/end_date). The range is the reliable way to answer a named period — for "the weekend" or "next Tuesday", the caller resolves today's date, then passes the exact dates rather than converting the period into a day count.
# 3-Day Forecast for 48.2208, 16.3738
Source: Open-Meteo (Europe/Vienna)
Wed 2026-07-22: 16–27°C — partlycloudy, wind up to 9 m/s
Thu 2026-07-23: 15–22°C — rainy, 4.2 mm (80% chance), wind up to 15 m/s
Fri 2026-07-24: 14–26°C — sunny, wind up to 8 m/s
Data sources & attribution
- GeoSphere Austria Dataset API — AROME forecast, INCA analysis/nowcast, C-LAEF ensemble. Data licensed under CC-BY 4.0. © GeoSphere Austria.
- Open-Meteo — worldwide forecast API. Data licensed under CC-BY 4.0. © Open-Meteo.
Both APIs are keyless and intended for non-commercial use. When you redistribute their data, keep the attribution.
Rate limits
The GeoSphere Dataset API allows 5 requests/second and 240 requests/hour. Each current/hourly call issues a small burst of concurrent requests; on an HTTP 429 the server retries once (when the API asks for a short wait) and otherwise returns a rate-limit notice — get_daily_forecast keeps working through Open-Meteo in that case. Open-Meteo has its own generous free-tier limits.
Development
# Install dependencies (creates .venv from the locked versions)
uv sync
# Lint & format
ruff check .
ruff format .
# Run unit tests
pytest -m "not integration"
# Run integration tests (hits the live GeoSphere + Open-Meteo APIs)
pytest -m integration
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。