metservice-nz

MetService NZ (MetOcean API)

Goal

Query point forecasts from the MetOcean API for New Zealand locations — weather, marine, and atmospheric data through a CLI that is easy for agents to script and easy for humans to scan.

Use this when

• A user asks about current or forecast weather in New Zealand
• A user wants temperature, rain, wind, or cloud data for an NZ location
• A user asks about marine conditions — wave height, swell, sea temperature
• A user wants a multi-day weather outlook for NZ
• A user asks about wind conditions for a specific NZ location

Do not use this for

• Weather outside New Zealand (API coverage is NZ/Pacific focused)
• Historical weather data (API provides forecasts, not archives)
• Long-range forecasts beyond ~7 days

Preferred workflow

1. Run scripts/cli.py with the narrowest subcommand that answers the task
2. Use default human output for direct answers
3. Use --json when another tool or agent needs machine-readable output
4. Use --location lat,lon for custom coordinates not in the named list

Setup

Requires a MetOcean API key (free tier available).

1. Sign up at https://forecast-v2.metoceanapi.com
2. Add to your .env:

   METOCEAN_API_KEY=your_key_here
   

3. The skill reads process.env.METOCEAN_API_KEY — no fallback, key is required.

CLI

Run with:

python3 skills/metservice-nz/scripts/cli.py <command> [flags]

now [location] [--json] [--location lat,lon]

Current conditions snapshot: temperature, wind, rain, humidity, pressure, cloud cover, visibility.

Examples:

python3 skills/metservice-nz/scripts/cli.py now auckland
python3 skills/metservice-nz/scripts/cli.py now wellington --json
python3 skills/metservice-nz/scripts/cli.py now --location -37.0,175.5

forecast [location] [--json] [--location lat,lon]

24-hour hourly forecast: time, temperature, rain, wind speed/direction, cloud cover.

Examples:

python3 skills/metservice-nz/scripts/cli.py forecast auckland
python3 skills/metservice-nz/scripts/cli.py forecast christchurch --json

daily [location] [--json] [--location lat,lon]

7-day daily summary: high/low temperature, total rain, average wind, dominant wind direction.

Examples:

python3 skills/metservice-nz/scripts/cli.py daily queenstown
python3 skills/metservice-nz/scripts/cli.py daily wellington --json

marine [location] [--json] [--location lat,lon]

Marine forecast: wave height, swell height, swell period, swell direction, sea temperature (where available).

Examples:

python3 skills/metservice-nz/scripts/cli.py marine auckland
python3 skills/metservice-nz/scripts/cli.py marine clevedon --json

wind [location] [--json] [--location lat,lon]

Detailed wind forecast: speed, gusts, direction — hourly for 24 hours.

Examples:

python3 skills/metservice-nz/scripts/cli.py wind wellington
python3 skills/metservice-nz/scripts/cli.py wind auckland

rain [location] [--json] [--location lat,lon]

Precipitation forecast: hourly rain rate for 24 hours.

Examples:

python3 skills/metservice-nz/scripts/cli.py rain christchurch
python3 skills/metservice-nz/scripts/cli.py rain queenstown --json

warnings [--json]

Scrape MetService severe weather warnings page for active alerts, watches, and advisories.

Examples:

python3 skills/metservice-nz/scripts/cli.py warnings
python3 skills/metservice-nz/scripts/cli.py warnings --json

cyclone [location] [--json] [--location lat,lon] [--hours n]

Cyclone-specific data: pressure trend, extreme wind, storm surge indicators. Fetches 48h by default. Shows current snapshot, pressure trend, peak conditions, and hourly table.

Examples:

python3 skills/metservice-nz/scripts/cli.py cyclone auckland
python3 skills/metservice-nz/scripts/cli.py cyclone wellington --hours 72
python3 skills/metservice-nz/scripts/cli.py cyclone clevedon --json

pressure [location] [--json] [--location lat,lon]

Barometric pressure trend — the #1 cyclone indicator. Shows 12h history + 12h forecast with rate of change and severity classification.

Examples:

python3 skills/metservice-nz/scripts/cli.py pressure auckland
python3 skills/metservice-nz/scripts/cli.py pressure wellington --json

watch [location] [--location lat,lon] [--interval minutes] [--json]

Continuous cyclone monitoring. Polls every 5 minutes (configurable) and alerts on significant changes. Shows pressure change between polls.

Examples:

python3 skills/metservice-nz/scripts/cli.py watch auckland
python3 skills/metservice-nz/scripts/cli.py watch wellington --interval 2
python3 skills/metservice-nz/scripts/cli.py watch clevedon --json

locations [--json]

List all named locations with coordinates.

Examples:

python3 skills/metservice-nz/scripts/cli.py locations
python3 skills/metservice-nz/scripts/cli.py locations --json

Named Locations

Name	Lat	Lon
clevedon	-36.9697	175.0752
auckland	-36.8485	174.7633
wellington	-41.2865	174.7762
christchurch	-43.5321	172.6362
queenstown	-45.0312	168.6626

Available API Variables

Atmospheric

• air.temperature.at-2m (K)
• air.humidity.at-2m (%)
• air.pressure.at-sea-level (Pa)
• air.visibility (m)
• cloud.cover (fraction 0-1)
• precipitation.rate (mm/hr)
• radiation.flux.downward.shortwave (W/m²)

Wind

• wind.speed.at-10m (m/s)
• wind.direction.at-10m (degrees)
• wind.speed.gust.at-10m (m/s)

Marine/Wave

• wave.height (m)
• wave.height.primary-swell (m)
• wave.height.wind-sea (m)
• wave.period.peak (s)
• wave.period.primary-swell.peak (s)
• wave.direction.peak (degrees)
• wave.direction.primary-swell.mean (degrees)
• sea.temperature.at-surface (K)

Resources

• CLI entrypoint: scripts/cli.py
• Live smoke test: scripts/smoke_test.py

Notes

• API key required — set METOCEAN_API_KEY in your .env
• Temperature returned in Kelvin, converted to Celsius in display
• Pressure returned in Pascals, converted to hPa in display
• Wind speed returned in m/s, converted to km/h in display
• Marine data may return null for inland or sheltered points
• Default output is human-readable; --json output is for chaining into other tools
• A location must always be specified — there is no default

Cyclone Tracking Tools

Data indicators, CLI commands, MetOcean variables, and official NZ and Pacific sources: references/cyclone-tracking.md