Geoscience Skills plugin
A collection of 39 skills and 8 workflows for geoscience: well logs, seismic, 3D models, inversion, geostatistics and climate.
- 82 ★ on the repository
- Last commit
- MIT
Minimal access
Why this level
- The skills give instructions for Python calculations while the user installs and runs libraries and data.
Install
- Works in
Manual install
npx skills add SteadfastAsArt/geoscience-skills --full-depthInteractive choice of skills and agents. Add `--list` to only list them.
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The skills give instructions for Python calculations while the user installs and runs libraries and data.
This is third-party code. Review the repository files before installing.
What it does
The collection teaches an agent to work with professional geoscience libraries. The `lasio` skill reads and edits LAS well logs, `obspy` processes seismic waveforms, `gempy` builds implicit geological models with faults and folds and `flopy` runs groundwater modelling. Other skills cover geostatistics, inversion, NetCDF climate data and geochemistry. Beyond 39 domain skills there are 8 end-to-end workflows, for example `well-log-evaluation`, `seismic-interpretation` and `climate-analysis`, and a discovery skill, `using-geoscience-skills`, that tells the agent what to pick. The Agent Skills format is shared by Codex, Claude Code, GitHub Copilot, Gemini CLI, Windsurf, OpenCode, Cline and Roo Code.
Who it is for. Geophysicists, petrophysicists and hydrogeologists who compute on Python and want an agent to take the routine: reading formats, calculations and plots.
Good fit when
- A SEG-Y file must be opened and its first traces shown.
- Water saturation is computed from well logs with the Archie equation.
- An electrical resistivity survey is inverted and a resistivity section plotted.
- A NetCDF climate file is reduced to monthly anomalies.
Not a fit when
- The task lies outside geoscience or a standard Python library covers it.
- You need commercial packages: GeoLime requires a vendor license.
- The result goes into a report with no specialist review.
Example request
Read this SEG-Y file, show the first 10 traces and compute the amplitude spectrum.Limitations
Installing a skill does not install the required Python libraries: `lasio`, `segyio` and others are installed separately with `pip`. Choosing a workflow does not pull in related skills automatically, so they are listed in the install command. Scientific runs are verified on Linux, and GeoLime works only with a vendor license. The `skills` installer makes symbolic links by default, and `--copy` gives copies. Do not copy the whole repository into a skill directory; copy each skill folder whole with its `references/` and `scripts/`. Bundled data keep their own licenses stated in the README.
How to disable. Run `npx skills list`, find the skill and remove it with the skills manager or by hand from the agent skills directory. Python libraries installed with `pip` are removed separately.
FAQ
How do I install only the skills I need?
Add the `--skill` flag with names, for example `--skill using-geoscience-skills lasio segyio --agent codex claude-code`. The `--full-depth` flag includes nested workflows.
Which data formats are supported?
LAS and DLIS for well logs, SEG-Y for seismic and NetCDF for climate. Each library has its own skill with details.
What does the using-geoscience-skills skill do?
It is the entry point: it tells the agent which skill or workflow fits the task so it does not have to scan all 48.
Does the collection work in Cursor?
The README lists Codex, Claude Code, GitHub Copilot, Gemini CLI, Windsurf, OpenCode, Cline, Roo Code and OpenClaw. Cursor is not named among the install targets.
How was it verified?
The README records Codex tasks for well logs and SEG-Y, installation of 48 skills on nine targets and automated checks, as of 25 October 2026.
Discussion
No questions or reviews yet.
Related
Pluginsopengis-skills gives an agent 120 skills on GIS, CAD, AI, IoT, 3D and C Sharp: GDAL, QGIS, GeoServer, PostGIS.
Around 200 skills for Claude Code on bioinformatics, chemistry and biostatistics, from RNA-seq analysis to molecular docking
17 skills for computational biology: RNA-seq, single-cell, protein structure, docking, literature and figures
A Google DeepMind skill collection for science tasks: access to biology and chemistry databases, literature search and structural biology
166 skills for scientific work: bioinformatics, cheminformatics, clinical data, geospatial analysis and 100+ databases
24 skills for numerical methods and materials science: the agent checks stability, meshes, convergence and cluster runs.
README in short
The README lists capabilities by domain, sample requests and installation through `npx skills`, whole or skill by skill. The verification section reports installs on nine targets and scientific tests.