Back to Skills

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.

2,728stars318forksUpdated 7/30/2026

Security Assessment

Safe(95/100)
Security Score95/100

About earth2studio-data-fetch

Earth2Studio Data Fetch guides users through downloading weather and climate data via Earth2Studio's data source APIs for specific variables and times. It solves the discoverability and compatibility problem of a large, evolving catalogue of data sources by identifying which source can provide the requested variables, verifying support through the lexicon system, and producing a working fetch script that outputs an xarray DataArray.

The skill treats live documentation and per-source lexicon files as the source of truth, fetching the relevant data source doc pages and lexicon definitions before recommending anything, because APIs and available variables change between releases. Its interaction protocol extracts the user's variables (mapping plain-language names to Earth2Studio names via the E2STUDIO_VOCAB), time range, data type (analysis/reanalysis versus forecast), lead time, region, and output format, then narrows candidate sources such as GFS, HRRR, IFS, ARCO/CDS/WB2 ERA5, and observational sources. It verifies each requested variable appears in the source's VOCAB dict and surfaces key differentiators like temporal coverage, spatial resolution, and update frequency. The skill is deliberately scoped to data fetching and explicitly excludes inference pipelines, model discovery, and installation.

It targets researchers, data scientists, and meteorologists working with Earth2Studio in Python. Typical use cases include pulling ERA5 reanalysis for a historical date, fetching operational GFS or HRRR fields for a region, and generating a reproducible xarray-based download script. Note the skill's own benchmark reports a FAIL verdict with a duplicate-content finding, so it is best treated as guidance to review.

FAQ

What does this skill do?

It helps identify compatible Earth2Studio data sources for requested weather/climate variables and times, verifies variable support via the lexicon, and produces a working Python fetch script that returns an xarray DataArray.

What are the prerequisites?

Earth2Studio installed (e.g. via uv pip install earth2studio), Python 3.10+, network access to remote data stores like GCS/S3/CDS, and for CDS-based sources a valid CDS API key configured in ~/.cdsapirc.

How does it confirm a variable is available from a source?

It fetches the source's lexicon file from the Earth2Studio repo and checks that the requested variable appears as a key in that source's VOCAB dict; if not, that source cannot provide it and another is tried.

What is it not meant for?

It is scoped to data fetching only. It should not be used for building inference pipelines, discovering or comparing models, or installation tasks.

How reliable is the published benchmark?

The bundled BENCHMARK.md records an overall FAIL verdict driven by a duplicate-content finding in SKILL.md and an author-format schema note; security scored 100%, but owners are advised to address findings and rerun evaluation.

All Files

8 files
BENCHMARK.md3.6 KB
View
evals/targets/eval_4_target.py1.3 KB
View
SKILL.md8.5 KB
View
evals/targets/eval_1_target.py1.2 KB
View
skill-card.md3.9 KB
View
evals/evals.json2.8 KB
View
evals/targets/eval_7_target.py1.3 KB
View
skill.oms.sig5.2 KB
View

Install earth2studio-data-fetch

Download and extract the skill files to your .claude/skills/ directory.

Quick Setup:

  1. Copy the skill folder to .claude/skills/
  2. Claude will automatically detect and use the skill

Repository

nvidia/skills