Glm runner
Skill cxcscmu/SkillLearnBench/skills/b1-one-shot-gemini-3-flash-preview/temperature-simulation/glm-runner
A skill to execute the GLM binary and manage its simulation output.From its SKILL.md
npx -y skills add cxcscmu/SkillLearnBench --skill glm-runnerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
SKILL.md
1.1 KB, 236 tokens by cl100k_base, as published. Nobody here has run it
GLM Runner
Overview
The GLM model is typically a standalone binary executable. Running it involves ensuring all input files (namelist, forcing, initialization) are correctly linked and accessible.
Setup
Ensure the glm binary is in your PATH.
Check its version with:
glm --version
Running the Simulation
GLM is usually run from the directory containing glm3.nml.
# Execute GLM
glm
Output Monitoring
GLM logs progress to standard output. Success is usually indicated by the model finishing its time steps.
Output is typically written to a NetCDF file, e.g., output.nc.
Troubleshooting
- Input Error: Check that all CSV forcing files (e.g., in
bcs/) have the columns specified inglm3.nml. - Config Error: Namelist errors in
glm3.nml(like typos or missing parameters) will cause GLM to crash or use default values. - Range Error: Some parameters have physical limits. If GLM crashes early, check your parameter values.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.