# Check Current Weather Across Japan

Weather forecasts do not show what nearby stations have just measured. This skill guides assistants through public JMA observations with station names and timestamps.

## Install

```bash
npx skillstore add tahodev/amedas-weather
```

## Metadata

- Status: approved
- Slug: tahodev-amedas-weather
- Skillstore revision: r1
- Version status: missing
- Tree hash: 0f6b81b7de712a63a96888cd0a929c23ad6af409e388ac9348a7f779171cb707
- Author: tahodev
- GitHub username: tahodev
- License: MIT
- Repository: https://github.com/tahodev/kurashi-skill/tree/09cb7f9eac85f0c85633cce08328a1a57854a94f/amedas-weather
- Ref: 7f17813fb6af1acc7d8fd3fb5f05480f5be52783
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, network, filesystem
- Quality score: 77
- Quality tier: bronze
- Public page: https://skillstore.pages.dev/skills/tahodev-amedas-weather
- Manifest: https://skillstore.pages.dev/api/skills/tahodev-amedas-weather/manifest

## Capabilities

- Retrieve the latest JMA observation timestamp and construct the corresponding observation download URL.
- Read available temperature, humidity, wind, rainfall, sunshine, and pressure values from station observations.
- Resolve station codes to names, coordinates, and elevations using the public station table.
- Find an approximately nearest station using a Python example and supplied coordinates.
- Report observation times in JST and identify the station used.
- Handle retrieval failures and missing measurements with explicit explanations and official fallback information.

## Use Cases

- Check Nearby Conditions: Read recent temperature, humidity, wind, and rainfall measurements for a nearby Japanese station before a commute.
- Summarize Local Measurements: Prepare a timestamped observation summary for a field team, with missing measurements clearly identified.
- Inspect Public Weather Data: Explore station metadata, measurement availability, and quality flags when preparing a weather data analysis.

## Prompt Templates

### Check One City

```
Show the latest available temperature and humidity for Tokyo. Include the station name and observation time in JST.
```

### Inspect Rain and Wind

```
Check the latest wind and one-hour rainfall near Osaka. Identify the station, units, observation time, and any unavailable measurements.
```

### Select a Nearby Station

```
Use coordinates 35.68, 139.77 to find an approximately nearest AMeDAS station. Report available weather measurements with the station name and JST timestamp.
```

### Compare Station Observations

```
Compare Tokyo and Yokohama observations at the latest shared timestamp. Include available temperature, humidity, wind, rainfall, quality flags, and station names.
```

## Limitations

- Provides observations, not forecasts, rain radar, or weather warnings.
- Measurements update every ten minutes, and available elements differ between stations.
- Requires network access and curl; the nearby-station example also requires Python 3.
- The proximity example uses squared coordinate differences, and the referenced shared response checklist is not included.

## Best Practices

- Use the retrieved latest timestamp, and include the station name and observation time in every summary.
- Check measurement availability and quality flags before interpreting values, especially for sensitive decisions.
- Run downloads in a dedicated working directory to avoid replacing existing amap.json or stations.json files.

## Anti Patterns

- Do not describe observations as forecasts or predictions of future conditions.
- Do not treat missing measurements or retrieval failures as zero values or clear weather.
- Do not present historical examples as current observations or omit rainfall accumulation periods.

## Security Audit

- Audited at: 2026-10-13T03:03:49.788\+00:00
- Summary: All 21 static findings are false positives in context: Markdown backticks, public JMA endpoints, a documentation link, and readable multilingual text. The documented commands retrieve weather data and process station coordinates locally; no evidence of credential access, exfiltration, destructive behavior, or prompt injection was found. This assessment covers the supplied SKILL.md, not the unavailable linked documentation.

## Stats

- Views: 0
- Downloads: 0
- Favorites: 0
- Popularity score: 0