submit-experiment

Automate SLURM experiment submission with structured naming and documentation artifacts.

16|4|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/umich-foreseer/lab-claude-config --skill submit-experiment
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: submit-experiment
Source: https://github.com/umich-foreseer/lab-claude-config/tree/main/shared/skills/submit-experiment
Command: npx skills add https://github.com/umich-foreseer/lab-claude-config --skill submit-experiment

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manual, error-prone submission of SLURM experiments across clusters is common without consistent naming and documentation.

Core Features & Use Cases

  • Enforces structured experiment naming to ensure traceability across runs.
  • Generates and attaches documentation artifacts for each submission, including run context and provenance.
  • Supports cross-cluster submission workflows with SSH orchestration and local/remote execution.

Quick Start

Provide the job type, run configuration or script path, and a brief purpose description to submit an experiment.

Frequently Asked Questions about submit-experiment

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I submit SLURM jobs with consistent naming conventions and documentation?

To submit SLURM jobs with consistent naming, you provide the job type, run configuration, and a brief purpose description. The automation then enforces structured experiment naming and generates documentation artifacts for traceable provenance.

Does this automation handle cross-cluster SLURM submission workflows?

Yes, cross-cluster SLURM submission workflows are supported using SSH orchestration. This allows you to execute local and remote jobs across multiple clusters while maintaining consistent experiment naming and documentation.

What is the best way to track provenance for multi-cluster SLURM experiments?

The best way to track provenance is by enforcing structured naming conventions and generating documentation artifacts. This captures run context and configuration details automatically during SLURM experiment submission for reproducible logging.

What do I need to provide to automate SLURM experiment submission?

You need to provide the job type, run configuration or script path, and a brief purpose description for the experiment. This input enables the system to parse project context and enforce reproducible logging.

Can I use this for reproducible logging of local and remote SLURM workloads?

Yes, you can use this for reproducible logging of local and remote SLURM workloads. It automates run config parsing and attaches documentation artifacts to each submission to ensure consistent provenance across clusters.

Why does manual SLURM job submission across clusters fail to maintain provenance?

Manual SLURM job submission across clusters fails because it is error-prone and lacks consistent naming and documentation. Automating the process enforces structured experiment naming and generates run context artifacts to resolve this.