skills.rest
kang-jaehyun
kang-jaehyunCOMMUNITY
Explore all skills by this author →
GitHub

submit

Submit machine-learning training jobs to a SLURM cluster with resource requirements.

Updated Aug 23, 2026
Search Tags:#cluster#experiments#machine-learning#slurm#job#submit#sbatch
One-click install
npx skills add https://github.com/kang-jaehyun/Mac-the-researcher --skill submit-kang-jaehyun
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: submit
Source: https://github.com/kang-jaehyun/Mac-the-researcher/tree/main/.claude/skills/submit
Command: npx skills add https://github.com/kang-jaehyun/Mac-the-researcher --skill submit-kang-jaehyun

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Submits machine-learning experiments to a SLURM cluster to automate training job launches. Supports selecting the target cluster, specifying the script, and defining resource requirements (GPUs, time, memory) across multiple clusters. Requires explicit confirmation before submission, uses submit_job to launch the job, and records metadata with record_experiment for later monitoring via /jobs.

Core Features & Use Cases

  • Parse user requests to identify target cluster, script path, and resource requirements (GPUs, time, memory).
  • Confirm planned submission with the user, including the sbatch command and estimated resources.
  • Submit the job using the submit_job MCP tool and track the job lifecycle.
  • Record experiment metadata with record_experiment, linking cluster, job ID, and key hyperparameters.
  • Provide status and results guidance via commands like /jobs and log inspection.

Quick Start

Submit a training job to your cluster by stating the target cluster, script path, and resource requirements.

Frequently Asked Questions about submit

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

FAQPage Schema
How do I submit machine-learning experiments to a SLURM cluster from a chat interface?▼

You can submit machine-learning experiments to a SLURM cluster by stating the target cluster, script path, and resource requirements. The Skill parses your request and requires explicit confirmation before launching the training job.

What resource requirements can I specify when submitting a SLURM job?▼

When submitting a SLURM job, you can specify resource requirements including GPUs, time limits, and memory. The Skill parses these parameters to configure your sbatch command before requesting explicit confirmation.

How does job confirmation work before launching a SLURM training job?▼

Before launching a SLURM training job, the Skill presents the planned sbatch command and estimated resources for your review. You must provide explicit confirmation before the job is actually submitted to the cluster.

Can I submit machine-learning jobs across multiple SLURM clusters?▼

Yes, you can submit machine-learning jobs across multiple SLURM clusters. The Skill supports selecting the target cluster, allowing you to direct your training job launches to the appropriate environment.

How is experiment metadata tracked after submitting a SLURM job?▼

After submitting a SLURM job, the Skill records experiment metadata linking the cluster, job ID, and key hyperparameters. This metadata is stored for later monitoring and status tracking via commands like /jobs.

What do I need to monitor a SLURM job after it has been submitted?▼

To monitor a SLURM job after submission, you can use the /jobs command and perform log inspection. The Skill records experiment metadata during submission to enable this later lifecycle tracking and results guidance.

Related Skills

Explore More Software Engineering Skills→
QuantumBFSQuantumBFS
60

setup-cluster

Generate standardized TOML profiles and verify SSH access for HPC clusters.

Official
Advanced
NVIDIANVIDIA
320

log-analysis

Diagnoses SLURM training job failures and recommends restart viability per cycle.

Official
Advanced
gabrielfruetgabrielfruet
3

remote-slurm-bootstrap

Submit and monitor Slurm benchmark jobs on remote HPC clusters.

Community
Advanced
umich-foreseerumich-foreseer
16

slurm-job

Generate Slurm sbatch scripts with partition, account, GPU, and validation directives.

Official
Intermediate