Agent Skill · NVIDIA NIM

earth2studio-data-fetch

Fetch weather/climate data via Earth2Studio data sources for specific variables and times. Do NOT use for inference pipelines, model discovery, or installation.

View SKILL.md on GitHub → Source repository Provider profile
Provider: NVIDIA NIM Path in repo: skills/earth2studio-data-fetch/SKILL.md

Skill body

Earth2Studio Data Fetch Skill

Purpose

Guide a user through downloading weather/climate data via Earth2Studio data source APIs. Identifies compatible sources by checking the lexicon, verifies variable support, and produces a working fetch script outputting an xarray DataArray.

Prerequisites

Instructions

You are helping a user download specific weather/climate data using Earth2Studio’s data source APIs. Your job is to identify which data source(s) can provide the requested variables, verify compatibility via the lexicon system, and produce a working fetch script.

Core principle: live docs and lexicon are the source of truth

Data source APIs, available variables, and the lexicon evolve between releases. Before recommending a data source or writing a fetch script:

  1. Fetch the relevant data source doc page to confirm the API signature and constructor arguments.
  2. Check the lexicon to verify the requested variable is supported by that data source.

Live doc references (fetch only what the user’s request requires):

Interaction protocol

Step 1. Understand the user’s request

Extract from what the user has said (ask follow-ups if needed, cap at 3 questions):

Step 2. Identify candidate data sources

Based on the request type, narrow candidates:

Analysis/reanalysis (historical state at a specific time):

Forecast (predictions from an initialization time with lead times):

Key differentiators to surface:

Step 3. Verify variable support via lexicon

This is critical. Each data source has a lexicon file that defines which E2Studio variables it can provide.

To verify:

  1. Fetch the source’s lexicon file from https://github.com/NVIDIA/earth2studio/blob/main/earth2studio/lexicon/<source>.py (e.g. gfs.py, hrrr.py, cds.py, arco.py, wb2.py)
  2. Check that the user’s requested variable(s) appear as keys in the source’s VOCAB dict
  3. If a variable is NOT in a source’s lexicon, that source cannot provide it — try another

The lexicon VOCAB maps Earth2Studio variable names → source-specific identifiers. If a variable key exists in the VOCAB, the source supports it.

Present the results clearly: “GFS supports t2m, u500, z850. HRRR also supports these but is limited to North America. ARCO (ERA5) supports all three and has data back to 1959.”

Step 4. Confirm data source selection with user

Present the viable options with tradeoffs:

Source Variables Coverage Resolution Time Range
… … … … …

Let the user pick. If there’s one obvious choice, recommend it and ask for confirmation.

Step 5. Generate fetch script

Write a Python script that uses the selected data source to fetch the requested data. The script structure depends on whether it’s an analysis or forecast source.

Analysis source pattern:

import datetime
from earth2studio.data import <SourceClass>

# Initialize data source
ds = <SourceClass>()

# Fetch data
# Analysis sources use: ds(time, variable) -> xr.DataArray
time = [datetime.datetime(YYYY, M, D, H)]  # or array of times
variable = ["var1", "var2"]  # E2Studio variable names

data = ds(time, variable)

Forecast source pattern:

import datetime
from earth2studio.data import <SourceClass>

# Initialize data source
ds = <SourceClass>()

# Forecast sources use: ds(time, lead_time, variable) -> xr.DataArray
time = [datetime.datetime(YYYY, M, D, H)]  # initialization time
lead_time = [datetime.timedelta(hours=H)]   # or array of lead times
variable = ["var1", "var2"]

data = ds(time, lead_time, variable)

Always fetch the specific data source’s API doc page to confirm the exact constructor arguments and call signature before writing the script — they can vary (some need auth tokens, cache paths, specific parameters).

Include in the script:

Step 6. Offer next steps

After delivering the script, mention:

Ownership and out-of-scope

Owns: identifying data sources for a user’s variable/time request, verifying variable support via lexicon, generating data fetch scripts, explaining analysis vs. forecast source differences.

Does not own: installation (earth2studio-install), model selection (earth2studio-discover), inference pipelines, custom data source creation (point to extend examples), data source authentication setup beyond what the docs describe.

Examples

Typical invocation:

“I need 500 hPa geopotential height and 2m temperature from ERA5 for January 1, 2020 at 00Z.”

The skill would:

  1. Map plain language → z500, t2m
  2. Check ARCO/CDS/WB2ERA5 lexicons for support
  3. Recommend ARCO (free, no API key) or CDS (official, needs key)
  4. Generate a fetch script using the selected source

Limitations

Troubleshooting

Error Cause Solution
KeyError: '<var>' Not in lexicon Check lexicon; try another source
FileNotFoundError / 404 Time not available Verify temporal coverage
CDS API timeout Queue congestion Retry or use ARCO for ERA5
ModuleNotFoundError Not installed uv pip install earth2studio
Empty DataArray Time/var mismatch Check datetime and variable name

Skill frontmatter

version: 0.16.0 license: Apache-2.0 metadata: {"author" => "NVIDIA Earth-2 Team", "tags" => ["earth2studio", "earth2", "python", "data-fetch", "weather-data", "xarray"]}