hubot-pollen
Version:
Retrieves the latest from the Pollen.com API
144 lines (104 loc) • 4.46 kB
Markdown
Retrieves the latest pollen forecast using Open-Meteo first, with fallback to Pollen.com when Open-Meteo has no usable pollen data.
[](http://badge.fury.io/js/hubot-pollen) [](https://github.com/stephenyeargin/hubot-pollen/actions/workflows/nodejs.yml)
In hubot project repo, run:
`npm install hubot-pollen --save`
Then add **hubot-pollen** to your `external-scripts.json`:
```json
[
"hubot-pollen"
]
```
Set the target location by providing a default search value (ZIP, city, or place text). Defaults to Nashville, TN for demonstration purposes.
| Environment Variable | Required? | Description |
| ----------------------------- | :-------: | -------------------------------------------------------------- |
| `HUBOT_POLLEN_ZIP` | No | Default location query (ZIP, city, or place) |
| `HUBOT_POLLEN_OLLAMA_ENABLED` | No | Set to `true` to register pollen tools with `hubot-ollama` |
## Commands
- `hubot pollen` - Show today's pollen levels for the default location query.
- `hubot pollen <zip>` - Show today's pollen levels for the given 5-digit ZIP.
- `hubot pollen <search>` - Show today's pollen levels for a geocoded location query.
## Data Source Strategy
1. Query Open-Meteo geocoding with the provided location string.
2. Query Open-Meteo air-quality pollen endpoint by coordinate.
3. If Open-Meteo pollen values are all null (or otherwise unusable), fallback to Pollen.com using a 5-digit ZIP when available.
Open-Meteo pollen values are only available for Europe during pollen season, so fallback is expected for many US locales.
Slack responses cite Open-Meteo in the attachment footer when that data source is used.
## Sample Interaction
```
user1>> hubot pollen
hubot>> Nashville, TN Pollen: 8.2 (Medium-High) - Alder, Juniper, Maple
user1>> hubot pollen London
hubot>> London, GB Pollen: Low intensity (1.7 grains/m^3 peak) - Grass 1.7, Birch 0.6
```
## NPM Module
https://www.npmjs.com/package/hubot-pollen
## Ollama Integration
This package can expose pollen tools to [Ollama](https://ollama.ai/) for use in LLM-powered conversations through the [hubot-ollama](https://github.com/stephenyeargin/hubot-ollama) package.
### Enabling Ollama Tools
To enable this feature, set the following environment variable:
```bash
HUBOT_POLLEN_OLLAMA_ENABLED=true
```
This requires hubot-ollama to be installed in your hubot repository. If hubot-ollama is not available, the tools will not be registered but the package will continue to function normally.
Retrieves pollen forecast data for a specified location.
Parameters:
- location (string, required): Location to get pollen forecast for. Can be a city name, city/state, city/country, or US ZIP code (for example, "Nashville, TN", "London", "37203").
Returns:
- Source used (open-meteo or pollen.com)
- Location label
- Coordinates when available
- Intensity and peak concentration
- Top pollen types when available
#### get_pollen_by_coordinates
Retrieves pollen forecast data for specific latitude and longitude coordinates.
Parameters:
- latitude (number, required): Latitude coordinate
- longitude (number, required): Longitude coordinate
Returns:
- Source used (open-meteo)
- Coordinates
- Intensity and peak concentration
- Top pollen types when available
### Example Tool Response Shapes
**`get_pollen_forecast`** — Open-Meteo result:
```json
{
"source": "open-meteo",
"location": "London, GB",
"coordinates": { "latitude": 51.5, "longitude": -0.1 },
"intensity": "Low",
"peak_grains_per_m3": 1.7,
"pollen_types": [
{ "type": "Grass", "peak": 1.7 },
{ "type": "Birch", "peak": 0.6 }
]
}
```
**`get_pollen_forecast`** — Pollen.com fallback result:
```json
{
"source": "pollen.com",
"location": "Nashville, TN",
"zip": "37206",
"index": 8.2,
"level": "Medium-High",
"triggers": ["Alder", "Juniper", "Maple"]
}
```
**`get_pollen_by_coordinates`** result:
```json
{
"source": "open-meteo",
"coordinates": { "latitude": 51.5, "longitude": -0.1 },
"intensity": "Low",
"peak_grains_per_m3": 1.7,
"pollen_types": [
{ "type": "Grass", "peak": 1.7 }
]
}
```