mcpField
S
spaceweather-mcp

spaceweather-mcp

Local
github.com/hoon1983Science & Education0Updated 18w ago

About spaceweather-mcp

spaceweather-mcp is an MCP server that provides AI agents with space weather and geomagnetic data from NOAA, NASA, USGS, and Kyoto. It covers current conditions, forecasts, alerts, aurora probability, and event history, with source-specific licensing documented.

Key features

  1. 0115 tools across four data sources
  2. 02Retrieve solar wind and geomagnetic conditions
  3. 03Check storm alerts and forecasts
  4. 04View aurora probabilities by location
  5. 05Query flares and space weather events
  6. 06Access observatory readings and Dst index

Use cases

  1. 01Check current geomagnetic and solar conditions
  2. 02Review storm forecasts and alerts
  3. 03Estimate local aurora probability
  4. 04Explore solar and geomagnetic event history

MCP support

Exposes 15 tools for live space weather and geomagnetic data to MCP clients. Kyoto Dst data is segregated and labeled for non-commercial use; the server is informational, not for operational safety decisions.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"hoon1983-spaceweather": {
"command": "uvx",
"args": [
"spaceweather-mcp"
]
}
}
}
Run commanduvx spaceweather-mcp

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
pypispaceweather-mcp0.1.0

Tools

15

Saved tool list · updated 2w ago

This tool takes no parameters.

Details

From the project README

spaceweather-mcp

A Model Context Protocol server for space weather and geomagnetic conditions — solar flares, the solar wind, geomagnetic storms (Kp/Dst), radiation storms, sunspot regions, aurora forecasts, and a catalogued event history — so an AI agent can answer "what's the sun/space weather doing right now, and will I see aurora?" from authoritative data.

Sibling to seismic-mcp: that one reconciles earthquakes across agencies; this one reports solar/geomagnetic conditions. They install side by side.

Status

15 tools across 4 data sources — all live and verified. Everything is U.S./NASA public domain except the Kyoto Dst index, which is non-commercial and clearly labeled (and kept segregated from the public-domain feeds).

Tools

NOAA SWPC (public domain)

ToolReturns
get_conditions_nowOne-call snapshot: Kp, Bz/Bt, wind speed/density, G/S/R, storm flag, summary
get_kp_indexObserved planetary Kp series (~1 week)
get_kp_forecastNOAA 3-day Kp forecast
get_solar_windL1 magnetic field + plasma (window: 1-day/7-day)
get_alertsRecent SWPC alerts/watches/warnings
get_noaa_scalesG/S/R scales for today + 3 days
get_27day_outlookDaily F10.7 flux, Ap, largest Kp for 27 days
get_solar_flaresRecent GOES X-ray flares (C/M/X class events)
get_radiation_stormCurrent proton flux + NOAA S-scale (radiation storm)
get_solar_regionsToday's sunspot regions + C/M/X flare probabilities
get_aurora_forecastOVATION aurora probability (pass lat/lon for a local %)

NASA DONKI (public domain) — get_space_weather_events (CME / GST / FLR / SEP / IPS / HSS event catalog)

USGS Geomagnetism (public domain) — list_observatories, get_observatory (magnetometer X/Y/Z/F)

Kyoto WDC (non-commercial) — get_dst (hourly disturbance storm-time index)

Run

uv run spaceweather-mcp        # stdio MCP server
uv run pytest                  # tests

Add to claude_desktop_config.json (after publishing to PyPI):

{ "mcpServers": { "spaceweather": { "command": "uvx", "args": ["spaceweather-mcp"] } } }

Data sources, attribution & licensing

  • NOAA SWPC (services.swpc.noaa.gov) — U.S. public domain. Credit NOAA SWPC.
  • NASA DONKI via CCMC (kauai.ccmc.gsfc.nasa.gov/DONKI) — U.S. public domain (no API key needed). Credit NASA/CCMC DONKI.
  • USGS Geomagnetism (geomag.usgs.gov) — U.S. public domain. Credit the U.S. Geological Survey. (Sourced directly from USGS, not via INTERMAGNET, to keep public-domain status.)
  • Kyoto WDC for Geomagnetism — Dst index, non-commercial use only: acknowledge "WDC for Geomagnetism, Kyoto", cite the Dst DOI; real-time values are provisional. The non-commercial restriction is documented in the get_dst tool description and the DstReading schema (both surfaced to MCP clients), and this feed is kept separate from the public-domain ones.

⚠️ Only the Kyoto Dst feed is non-commercial. It is segregated, and its non-commercial terms are stated in the tool/schema docs the client sees, so a downstream commercial user is not silently bound by those terms.

Deliberately not included: INTERMAGNET and SILSO sunspot number (both CC BY-NC — would taint the otherwise commercially-usable server); JPL Horizons ephemeris (different domain). F10.7 is available via get_27day_outlook; GFZ Hp30/Hp60 high-cadence indices (CC BY 4.0) are a possible future add.

Safety

Informational only. Not for operational decisions affecting power grids, aviation, satellites, or human safety — consult official NOAA SWPC products for those. Real-time values are preliminary and routinely revised.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.