run-experiment

Installation
SKILL.md

Run Experiment

Submit an ML experiment to a compute environment — local machine, SLURM HPC (Ibex, UW, etc.), or RunAI/Kubernetes (EPFL).

Generates a reproducible job script in jobs/ that is committed alongside the code, then provides the exact submit command to run.

Pair this skill with research-project-memory when a launched job should be linked to a planned experiment, evidence item, worktree, or project action. Pair this skill with project toolchain gates when generated job scripts should be checked before launch. Use shellcheck and shfmt when available, but do not require the user to install them just to generate a draft script.

Terminology:

  • local: the machine where the agent is running, usually the user's Mac
  • Git remote: GitHub/GitLab remote used for code sync, such as origin
  • server: SSH/HPC/RunAI execution environment such as quest, ibex-vscode, or epfl-haas

When launching on SLURM or RunAI, call it a server run rather than a remote run unless referring to the Git remote.

Skill Directory Layout

Related skills
Installs
32
GitHub Stars
4
First Seen
Apr 21, 2026