Ecmwf Observations
Retrieves weather station observations (SYNOP and METAR — 2 m temperature, dew point, humidity, wind, pressure, precipitation) from ECMWF's MARS archive, with the departures from ECMWF's first guess and analysis and QC flags — load it first for any request for observed or measured weather at a station or airport, past observations, a WMO station id or ICAO code, or comparing a forecast with what was observed. Use when a task mentions observations, station data, SYNOP, METAR, BUFR or ODB feedback, verifying or scoring a forecast against stations (bias, MAE, RMSE), observed heat index or humidex
Install
/plugin marketplace add ecmwf/ecmwf-weather-skillsThe repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.
npx skills add ecmwf/ecmwf-weather-skills --skill ecmwf-observationsgit clone --depth 1 https://github.com/ecmwf/ecmwf-weather-skills.git /tmp/ecmwf-weather-skills
mkdir -p ~/.claude/skills
cp -r /tmp/ecmwf-weather-skills/plugins/ecmwf-weather/skills/ecmwf-observations ~/.claude/skills/ecmwf-observationsPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# ECMWF station observations ## Contents - Sources (line 35) - Workflow (line 49) - Commands (line 68) - Verify a forecast against observations (line 94) - Thermal-comfort indices (line 113) - Processing and plots (line 121) - Licence and attribution (line 128) - Feedback (line 137) - When blocked (line 144) - References — `references/codes.md` (report types, varno, QC flags, missing values), `references/requests.md` (MARS request shapes, sizes, latency) Use ECMWF sources only. The one exception is NOAA Aviation Weather (aviationweather.gov): ICAO station metadata, and the latest METARs **only when the user asks for METARs or has no MARS observation rights and agrees** — always labelled "NOAA Aviation Weather (non-ECMWF)". Never use any other weather API or website. ## Sources | Need | Route | |---|---| | Observed values at a station, any past period, with QC | `obs.py series` — ODB feedback (default) | | Every report as received (incl. ones the analysis did not use) | `obs.py series --raw` — BUFR | | The latest METAR right now (user asked) | `obs.py series --station ICAO --latest-metar` — NOAA, non-ECMWF | | A forecast to compare with | `ecmwf-open-data` (`odpoint.py`) or `ecmwf-polytope` (`ptpoint.py`) | Feedback reaches MARS ~10 h after each 6-hour window (00/06/12/18 UTC, ±3 h); BUFR ~2 days. Feedback holds stations ECMWF's analysis received, with `fg_depar` (observation minus first
Details
- Skill name
ecmwf-observations- Repository
- ecmwf/ecmwf-weather-skills
- Path
plugins/ecmwf-weather/skills/ecmwf-observations/SKILL.md- License
- Apache-2.0
- Repo updated
- 8 November 2026
FAQ
What does the Ecmwf Observations skill do?
Retrieves weather station observations (SYNOP and METAR — 2 m temperature, dew point, humidity, wind, pressure, precipitation) from ECMWF's MARS archive, with the departures from ECMWF's first guess and analysis and QC flags — load it first for any request for observed or measured weather at a station or airport, past observations, a WMO station id or ICAO code, or comparing a forecast with what was observed. Use when a task mentions observations, station data, SYNOP, METAR, BUFR or ODB feedback, verifying or scoring a forecast against stations (bias, MAE, RMSE), observed heat index or humidex
How do I install the Ecmwf Observations skill?
Claude Code plugin: /plugin marketplace add ecmwf/ecmwf-weather-skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Ecmwf Observations popular?
It lives in ecmwf/ecmwf-weather-skills, which has 23 GitHub stars and 0 forks.
Related skills
Ecmwf Cds Ads
23ecmwf/ecmwf-weather-skills
Answers questions about past weather at a place — how hot, cold, wet or windy it was on a day, month or year since 1940 — from ERA5 in seconds; load it before answering from memory or searching the web (the bundled ECMWF MCP server has no ERA5). Finds, validates and downloads data from the Copernicus Climate Data Store (CDS) and Atmosphere Data Store (ADS), operated by ECMWF — ERA5 and ERA5-Land reanalysis (hourly point series, climate normals), the CARRA (Arctic, 2.5 km) and CERRA / CERRA-Land (Europe, 5.5 km) regional reanalyses, seasonal forecasts, and CAMS air-quality and atmospheric-compo
Ecmwf Destine
23ecmwf/ecmwf-weather-skills
Handles any Destination Earth question — load it first; it prepares requests by running scripts/destine.py. Accesses Destination Earth (DestinE) Digital Twin data — the Climate Change Adaptation Digital Twin (Climate DT, multi-decadal IFS-FESOM, IFS-NEMO and ICON simulations and SSP3-7.0 projections, hourly or as monthly means), the Weather-Induced Extremes Digital Twin (Extremes DT, high-resolution forecasts) and On-Demand Extremes — through the DestinE Polytope service. Use when a task mentions Destination Earth, DestinE, Digital Twins, Climate DT, Extremes DT, nextGEMS, the DestinE Platform
Ecmwf Earthkit
23ecmwf/ecmwf-weather-skills
Processes, computes and plots ECMWF weather and climate data with ECMWF's earthkit components — load it before writing ANY numpy, xarray, pandas, matplotlib or cartopy code for weather data, and for any local GRIB (.grib, .grib2) or NetCDF file. Covers reading and selecting fields (earthkit-data), ECMWF-styled maps, multi-panel figures, meteograms and plumes (earthkit-plots), ensemble mean/spread/percentiles, daily and monthly statistics, de-accumulation, climatologies, anomalies and area or country means (earthkit-transforms), wind, humidity, dewpoint, potential temperature, wet bulb, EFI and
Enterprise Capacity Planning
45j4flmao/agent-skills
Use when the user asks about capacity planning, forecasting infrastructure growth, headroom, peak vs sustained load, hardware procurement lead times or cost per unit of work. Produces a capacity model and multi-quarter roadmap. Not for cloud bill optimization (use devops-finops) or autoscaling config (use devops-kubernetes-autoscaling).
Distributed Training
45j4flmao/agent-skills
Use when the user asks about multi-GPU training, DDP, FSDP, ZeRO, DeepSpeed, torchrun, tensor/pipeline parallel, all-reduce or NCCL slowness. Produces a parallelism plan with per-GPU memory math, launch command and config. Not for fabric tuning (use rdma-roce) or experiment logging (use ml-experiment-tracking).
Enterprise Vendor Management
45j4flmao/agent-skills
Use when the user asks about selecting a vendor or SaaS supplier, an RFP, vendor risk assessment, contract and SLA negotiation, vendor scorecards or exit planning. Produces vendor evaluation, risk and performance management plans. Not for build vs buy financials (use planning-cost-benefit) or hiring people (use management-hiring).
Browse the most popular Claude skills, all productivity & notes skills, or write your own with the skills generator.