.codex-plugin— mcp server

<div align="center"> <h1>@cyanheads/noaa-spaceweather-mcp-server</h1> <p><b>Query NOAA SWPC space weather: geomagnetic storm scales, Kp index, aurora forecasts, solar wind, solar activity, and alerts via MCP. STDIO or Streamable HTTP.</b> <div>6 Tools</div> </p> </div>

<div align="center">

</div>

<div align="center">

</div>

<div align="center">

Public Hosted Server: https://noaa-spaceweather.caseyjhand.com/mcp

</div>

Tools

Six tools covering the full NOAA SWPC space weather surface — from a quick status heartbeat to deep solar and geomagnetic data:

noaa_spaceweather_get_conditions

The heartbeat tool. Use it first — it composes storm scales + current Kp into a single snapshot with a plain-language summary.

• Returns today's R/S/G storm scales plus a 3-day forecast
• Current Kp with G-scale equivalent and aurora-visibility latitude guidance
• Plain-language summary: "Quiet conditions" vs. "G2 moderate geomagnetic storm in progress"
• Data sourced from noaa-scales.json + noaa-planetary-k-index.json

noaa_spaceweather_get_kp_index

Planetary K-index time series for geomagnetic activity tracking.

• window_days parameter (1–7, default 1) — recent 3-hour observed values
• Each record includes Kp value, G-scale equivalent, aurora-latitude guidance, and observation time
• Separate 3-day forecast series with predicted Kp and NOAA G-scale labels

noaa_spaceweather_get_aurora_forecast

OVATION model aurora probability at 1° resolution, updated every ~5 minutes.

• Without coordinates: global metadata — grid point count, global peak probability, peak region
• With coordinates (latitude, longitude): nearest-grid-point lookup, minimum Kp required at that latitude, and a plain-language verdict ("Good aurora chance (42%) — Kp≥6 needed at this latitude")
• Covers both hemispheres; coordinates are geographic (WGS84)

noaa_spaceweather_get_solar_wind

DSCOVR satellite real-time solar wind plasma and magnetic field.

• window_hours parameter (1–168, default 3) — time series sliced client-side from 7-day feed
• Plasma: speed (km/s), proton density (n/cm³), temperature
• Bz component (southward Bz = storm driver) surfaced prominently in output and format
• Bt (total field magnitude) and GSM vector components

noaa_spaceweather_get_solar_activity

Solar flare and radiation storm picture from GOES and SWPC probabilities.

• GOES primary X-ray flux in the 0.1–0.8 nm band (R-scale driver)
• 3-day C/M/X flare-class probabilities
• Active solar regions with per-region flare and proton probabilities when include_regions: true (default)
• Solar radiation storm (S-scale) level from ≥10 MeV integral proton flux

noaa_spaceweather_get_alerts

Active SWPC alerts, watches, and warnings from the products/alerts.json feed.

• Structured product metadata: type (Watch/Warning/Alert/Summary), issue time, and ISO 8601 validity window parsed from the message body (Valid From/To, Now Valid Until, or Begin/End Time)
• Parsed severity from product code (G/R/S-scale, K-index level)
• Filtered to active-only by default (active_only: true); set false for recent history
• Full raw message text preserved for downstream display

Features

Built on @cyanheads/mcp-ts-core:

• Declarative tool definitions — single file per tool, framework handles registration and validation
• Unified error handling — handlers throw, framework catches, classifies, and formats
• Pluggable auth: none, jwt, oauth
• Swappable storage backends: in-memory, filesystem, Supabase, Cloudflare KV/R2/D1
• Structured logging with optional OpenTelemetry tracing
• STDIO and Streamable HTTP transports

Space weather domain:

• All SWPC feeds are public and keyless — no API keys required
• Single SpaceWeatherService wraps all six NOAA SWPC JSON feeds with fetchWithTimeout + withRetry
• Heterogeneous feed normalization: array-of-arrays (solar wind), keyed objects (storm scales), coordinate triples (OVATION)
• NOAA scale interpretation: raw Kp 6 → "G2 moderate storm — aurora possible to ~55° geomagnetic latitude"
• Feed freshness surfaced per-response: solar wind updates ~1 min, aurora ~5 min, Kp 3-hour intervals

Agent-friendly output:

• Observed timestamps on every response so agents can reason about data freshness
• Plain-language summaries and verdicts alongside raw values — agents can display or reason without re-interpreting indices
• Bz component surfaced as a first-class field in solar wind output (southward Bz = primary storm driver)
• Typed error contracts with recovery hints: feed_unavailable → "Retry in 30–60 s"

Getting started

Public Hosted Instance

A public instance is available at https://noaa-spaceweather.caseyjhand.com/mcp — no installation required. Point any MCP client at it via Streamable HTTP:

{
  "mcpServers": {
    "noaa-spaceweather-mcp-server": {
      "type": "streamable-http",
      "url": "https://noaa-spaceweather.caseyjhand.com/mcp"
    }
  }
}

Self-Hosted / Local

Add the following to your MCP client configuration file.

{
  "mcpServers": {
    "noaa-spaceweather": {
      "type": "stdio",
      "command": "bunx",
      "args": ["@cyanheads/noaa-spaceweather-mcp-server@latest"],
      "env": {
        "MCP_TRANSPORT_TYPE": "stdio",
        "MCP_LOG_LEVEL": "info"
      }
    }
  }
}

Or with npx (no Bun required):

{
  "mcpServers": {
    "noaa-spaceweather": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@cyanheads/noaa-spaceweather-mcp-server@latest"],
      "env": {
        "MCP_TRANSPORT_TYPE": "stdio",
        "MCP_LOG_LEVEL": "info"
      }
    }
  }
}

Or with Docker:

{
  "mcpServers": {
    "noaa-spaceweather": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-e", "MCP_TRANSPORT_TYPE=stdio",
        "ghcr.io/cyanheads/noaa-spaceweather-mcp-server:latest"
      ]
    }
  }
}

For Streamable HTTP, set the transport and start the server:

MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 bun run start:http
# Server listens at http://localhost:3010/mcp

Prerequisites

• Bun v1.3.0 or higher (or Node.js v24+).
• No API keys required — all SWPC feeds are public.

Installation

1. Clone the repository:

git clone https://github.com/cyanheads/noaa-spaceweather-mcp-server.git

1. Navigate into the directory:

cd noaa-spaceweather-mcp-server

1. Install dependencies:

bun install

1. Configure environment:

cp .env.example .env
# edit .env if needed — all defaults work out of the box

Configuration

No domain-specific API keys are required. See .env.example for the full list of optional overrides.

Running the server

Local development

• Build and run:

  # One-time build
  bun run rebuild

  # Run the built server
  bun run start:stdio
  # or
  bun run start:http

• Run checks and tests:

  bun run devcheck   # Lint, format, typecheck, security
  bun run test       # Vitest test suite
  bun run lint:mcp   # Validate MCP definitions against spec

Docker

docker build -t noaa-spaceweather-mcp-server .
docker run --rm -p 3010:3010 noaa-spaceweather-mcp-server

The Dockerfile defaults to HTTP transport, stateless session mode, and logs to /var/log/noaa-spaceweather-mcp-server. OpenTelemetry peer dependencies are installed by default — build with --build-arg OTEL_ENABLED=false to omit them.

Project structure

Development guide

See CLAUDE.md/AGENTS.md for development guidelines and architectural rules. The short version:

• Handlers throw, framework catches — no try/catch in tool logic
• Use ctx.log for request-scoped logging, ctx.state for tenant-scoped storage
• Register new tools via the barrel in src/mcp-server/tools/definitions/index.ts
• Wrap external API calls: validate raw → normalize to domain type → return output schema; never fabricate missing fields

Contributing

Issues and pull requests are welcome. Run checks and tests before submitting:

bun run devcheck
bun run test

License

Apache-2.0 — see LICENSE for details.