run-job

Automate SLURM job execution from sbatch generation to artifact upload.

5|Updated Mar 27, 2026

Search Tags:#automation #monitoring #experiment #slurm #job #sbatch

Ready to run

npx skills add https://github.com/Zayne-sprague/Dr-Claude-Code --skill run-job

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: run-job
Source: https://github.com/Zayne-sprague/Dr-Claude-Code/tree/main/.claude/skills/run-job
Command: npx skills add https://github.com/Zayne-sprague/Dr-Claude-Code --skill run-job

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the end-to-end lifecycle of running experimental jobs: from generating the sbatch script to submitting, monitoring, handling artifacts, resuming after failure, and reporting results.

Core Features & Use Cases

Write and submit sbatch scripts for SLURM-based experiments and canaries.
Monitor job progress, handle retries, and manage artifacts across the lifecycle.
Pause, resume, and recover from failures with checkpoints, ensuring reproducible runs.

Quick Start

Provide an experiment design and target cluster, and I will generate, submit, and monitor the sbatch job end-to-end.

Frequently Asked Questions about run-job

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

FAQPage Schema

How do I automate SLURM job execution from experiment design to results?▼

You can automate SLURM job execution by providing an experiment design and target cluster to generate, submit, and monitor sbatch jobs end-to-end. This enforces checkpointing, logs output, uploads artifacts, and supports resume on failure.

What is the best way to handle SLURM job failures and resume from checkpoints?▼

Handling SLURM job failures involves implementing checkpointing within your sbatch scripts and monitoring job progress. You can resume experiments from checkpoints automatically upon failure, ensuring reproducible runs without losing computational progress.

Can I use sbatch scripts for experiments on RunPod or local compute?▼

Yes, sbatch scripts generated for experiment designs can be applied across SLURM, RunPod, or local compute targets. The orchestration process fulfills hard rules like writing sbatch from templates, ensuring checkpointing, and uploading artifacts regardless of the target environment.

How do I generate sbatch scripts from experiment templates for canary specifications?▼

Generating sbatch scripts from templates requires providing your experiment design and canary specifications. The system writes the sbatch script, submits the job, monitors progress, handles retries, and manages artifacts across the entire lifecycle.

Does SLURM job orchestration ensure reproducible experiment runs with artifact management?▼

SLURM job orchestration ensures reproducible experiment runs by enforcing checkpointing, logging output, and uploading artifacts. You can pause, resume, and recover from failures with checkpoints throughout the job lifecycle.