KosmoKrator

data

Open-Meteo Lua API for KosmoKrator Agents

Agent-facing Lua documentation and function reference for the Open-Meteo KosmoKrator integration.

Lua Namespace

Agents call this integration through app.integrations.open_meteo.*. Use lua_read_doc("integrations.open-meteo") inside KosmoKrator to discover the same reference at runtime.

Call Lua from the Headless CLI

Use kosmo integrations:lua when a shell script, CI job, cron job, or another coding CLI should run a deterministic Open-Meteo workflow without starting an interactive agent session.

Inline Lua call 
kosmo integrations:lua --eval 'dump(app.integrations.open_meteo.weather_forecast({}))' --json
Read Lua docs headlessly 
kosmo integrations:lua --eval 'print(docs.read("open-meteo"))' --json
kosmo integrations:lua --eval 'print(docs.read("open-meteo.weather_forecast"))' --json

Workflow file

Put repeatable logic in a Lua file, then execute it with JSON output for the calling process.

workflow.lua 
local open_meteo = app.integrations.open_meteo
local result = open_meteo.weather_forecast({})

dump(result)
Run the workflow 
kosmo integrations:lua workflow.lua --json
kosmo integrations:lua workflow.lua --force --json
Namespace note. integrations:lua exposes app.integrations.open_meteo, app.mcp.*, docs.*, json.*, and regex.*. Use app.integrations.open_meteo.default.* or app.integrations.open_meteo.work.* when you configured named credential accounts.

MCP-only Lua

If the script only needs configured MCP servers and does not need Open-Meteo, use the narrower mcp:lua command.

MCP Lua command 
# Use mcp:lua for MCP-only scripts; use integrations:lua for this integration namespace.
kosmo mcp:lua --eval 'dump(mcp.servers())' --json

Agent-Facing Lua Docs

This is the rendered version of the full Lua documentation exposed to agents when they inspect the integration namespace.

Open-Meteo

Open-Meteo tools are exposed under app.integrations.open_meteo. The integration uses public Open-Meteo APIs and requires no credentials for standard non-commercial use.

Coverage

  • open_meteo_forecast - https://api.open-meteo.com/v1/forecast
  • open_meteo_model_forecast - https://api.open-meteo.com/v1/{model_endpoint}
  • open_meteo_historical_weather - https://archive-api.open-meteo.com/v1/archive
  • open_meteo_historical_forecast - https://historical-forecast-api.open-meteo.com/v1/forecast
  • open_meteo_ensemble - https://ensemble-api.open-meteo.com/v1/ensemble
  • open_meteo_seasonal_forecast - https://seasonal-api.open-meteo.com/v1/seasonal
  • open_meteo_climate - https://climate-api.open-meteo.com/v1/climate
  • open_meteo_marine - https://marine-api.open-meteo.com/v1/marine
  • open_meteo_air_quality - https://air-quality-api.open-meteo.com/v1/air-quality
  • open_meteo_satellite_radiation - https://satellite-api.open-meteo.com/v1/satellite-radiation
  • open_meteo_flood - https://flood-api.open-meteo.com/v1/flood
  • open_meteo_elevation - https://api.open-meteo.com/v1/elevation
  • open_meteo_geocoding_search - https://geocoding-api.open-meteo.com/v1/search
  • open_meteo_geocoding_get - https://geocoding-api.open-meteo.com/v1/get

Usage Notes

Pass official Open-Meteo query parameters as top-level tool arguments. For less common or newly added parameters, pass them inside query; top-level arguments override matching keys in query. Array values are encoded as comma-separated lists, matching Open-Meteo’s documented variable selection style.

Most weather endpoints require latitude and longitude. Historical and climate endpoints also require start_date and end_date. Climate calls require models and daily. Geocoding search uses name, while geocoding get uses id.

open_meteo_model_forecast accepts model_endpoint, for example gfs, ecmwf, icon, meteofrance, ukmo, jma, kma, metno, gem, bom, cma, knmi, dmi, or meteoitalia.

Examples

local forecast = app.integrations.open_meteo.open_meteo_forecast({
  latitude = 52.52,
  longitude = 13.41,
  hourly = { "temperature_2m", "precipitation" },
  daily = { "temperature_2m_max", "temperature_2m_min" },
  timezone = "Europe/Berlin"
})

local berlin = app.integrations.open_meteo.open_meteo_geocoding_search({
  name = "Berlin",
  count = 5,
  language = "en"
})

local air = app.integrations.open_meteo.open_meteo_air_quality({
  latitude = 52.52,
  longitude = 13.41,
  hourly = { "pm10", "pm2_5", "us_aqi" }
})

Responses are decoded Open-Meteo JSON responses. Open-Meteo may return a JSON object with error = true and reason; the integration converts that into a tool error.

Raw agent markdown 
# Open-Meteo

Open-Meteo tools are exposed under `app.integrations.open_meteo`. The integration uses public Open-Meteo APIs and requires no credentials for standard non-commercial use.

## Coverage

- `open_meteo_forecast` - `https://api.open-meteo.com/v1/forecast`
- `open_meteo_model_forecast` - `https://api.open-meteo.com/v1/{model_endpoint}`
- `open_meteo_historical_weather` - `https://archive-api.open-meteo.com/v1/archive`
- `open_meteo_historical_forecast` - `https://historical-forecast-api.open-meteo.com/v1/forecast`
- `open_meteo_ensemble` - `https://ensemble-api.open-meteo.com/v1/ensemble`
- `open_meteo_seasonal_forecast` - `https://seasonal-api.open-meteo.com/v1/seasonal`
- `open_meteo_climate` - `https://climate-api.open-meteo.com/v1/climate`
- `open_meteo_marine` - `https://marine-api.open-meteo.com/v1/marine`
- `open_meteo_air_quality` - `https://air-quality-api.open-meteo.com/v1/air-quality`
- `open_meteo_satellite_radiation` - `https://satellite-api.open-meteo.com/v1/satellite-radiation`
- `open_meteo_flood` - `https://flood-api.open-meteo.com/v1/flood`
- `open_meteo_elevation` - `https://api.open-meteo.com/v1/elevation`
- `open_meteo_geocoding_search` - `https://geocoding-api.open-meteo.com/v1/search`
- `open_meteo_geocoding_get` - `https://geocoding-api.open-meteo.com/v1/get`

## Usage Notes

Pass official Open-Meteo query parameters as top-level tool arguments. For less common or newly added parameters, pass them inside `query`; top-level arguments override matching keys in `query`. Array values are encoded as comma-separated lists, matching Open-Meteo's documented variable selection style.

Most weather endpoints require `latitude` and `longitude`. Historical and climate endpoints also require `start_date` and `end_date`. Climate calls require `models` and `daily`. Geocoding search uses `name`, while geocoding get uses `id`.

`open_meteo_model_forecast` accepts `model_endpoint`, for example `gfs`, `ecmwf`, `icon`, `meteofrance`, `ukmo`, `jma`, `kma`, `metno`, `gem`, `bom`, `cma`, `knmi`, `dmi`, or `meteoitalia`.

## Examples

```lua
local forecast = app.integrations.open_meteo.open_meteo_forecast({
  latitude = 52.52,
  longitude = 13.41,
  hourly = { "temperature_2m", "precipitation" },
  daily = { "temperature_2m_max", "temperature_2m_min" },
  timezone = "Europe/Berlin"
})

local berlin = app.integrations.open_meteo.open_meteo_geocoding_search({
  name = "Berlin",
  count = 5,
  language = "en"
})

local air = app.integrations.open_meteo.open_meteo_air_quality({
  latitude = 52.52,
  longitude = 13.41,
  hourly = { "pm10", "pm2_5", "us_aqi" }
})
```

Responses are decoded Open-Meteo JSON responses. Open-Meteo may return a JSON object with `error = true` and `reason`; the integration converts that into a tool error.
Metadata-derived Lua example 
local result = app.integrations.open_meteo.weather_forecast({})
print(result)

Functions

weather_forecast Read

Get standard Open-Meteo forecast data.

Lua path
app.integrations.open_meteo.weather_forecast
Full name
open-meteo.open_meteo_forecast
ParameterTypeRequiredDescription
No parameters.
model_forecast Read

Get forecast data from a specific Open-Meteo model endpoint.

Lua path
app.integrations.open_meteo.model_forecast
Full name
open-meteo.open_meteo_model_forecast
ParameterTypeRequiredDescription
No parameters.
historical_weather Read

Get historical weather archive data.

Lua path
app.integrations.open_meteo.historical_weather
Full name
open-meteo.open_meteo_historical_weather
ParameterTypeRequiredDescription
No parameters.
historical_forecast Read

Get past forecast model runs.

Lua path
app.integrations.open_meteo.historical_forecast
Full name
open-meteo.open_meteo_historical_forecast
ParameterTypeRequiredDescription
No parameters.
ensemble_forecast Read

Get ensemble forecast data.

Lua path
app.integrations.open_meteo.ensemble_forecast
Full name
open-meteo.open_meteo_ensemble
ParameterTypeRequiredDescription
No parameters.
seasonal_forecast Read

Get seasonal and sub-seasonal forecast data.

Lua path
app.integrations.open_meteo.seasonal_forecast
Full name
open-meteo.open_meteo_seasonal_forecast
ParameterTypeRequiredDescription
No parameters.
climate_projections Read

Get climate projection data.

Lua path
app.integrations.open_meteo.climate_projections
Full name
open-meteo.open_meteo_climate
ParameterTypeRequiredDescription
No parameters.
marine_forecast Read

Get marine weather and wave forecasts.

Lua path
app.integrations.open_meteo.marine_forecast
Full name
open-meteo.open_meteo_marine
ParameterTypeRequiredDescription
No parameters.
air_quality Read

Get air quality forecasts.

Lua path
app.integrations.open_meteo.air_quality
Full name
open-meteo.open_meteo_air_quality
ParameterTypeRequiredDescription
No parameters.
satellite_radiation Read

Get satellite radiation data.

Lua path
app.integrations.open_meteo.satellite_radiation
Full name
open-meteo.open_meteo_satellite_radiation
ParameterTypeRequiredDescription
No parameters.
flood_forecast Read

Get flood and river discharge forecasts.

Lua path
app.integrations.open_meteo.flood_forecast
Full name
open-meteo.open_meteo_flood
ParameterTypeRequiredDescription
No parameters.
elevation Read

Get elevation for coordinates.

Lua path
app.integrations.open_meteo.elevation
Full name
open-meteo.open_meteo_elevation
ParameterTypeRequiredDescription
No parameters.
geocoding_get Read

Resolve a geocoding location ID.

Lua path
app.integrations.open_meteo.geocoding_get
Full name
open-meteo.open_meteo_geocoding_get
ParameterTypeRequiredDescription
No parameters.