| name | compute-env-setup |
| description | Set up a reproducible Feynman compute environment for research jobs. Use when a task needs Python/R packages, GPU libraries, containers, Modal, SSH, caches, or managed model runtime setup. |
Compute Environment Setup
Use this skill before running a research job that needs a nontrivial runtime.
Workflow:
- Identify the job type, hardware need, data size, package stack, secrets, and expected artifacts.
- Choose the smallest working route: local notebook runtime, project virtualenv/conda, Modal, SSH/Slurm, or managed endpoint.
- Verify credentials and CLIs through Feynman Settings or environment status without printing secret values.
- Write an environment note containing package versions, install commands, cache paths, hardware, and failure modes.
- Run a tiny smoke job before the expensive job and save the smoke logs.
The done state is a recorded, reproducible environment plus a successful smoke or a precise missing dependency.