lab-cluster-1

Manage PJLAB lab cluster SSH, job submission, and service workflows.

7|1|Updated May 9, 2026
One-click install
npx skills add https://github.com/black-yt/skills --skill lab-cluster-1
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: lab-cluster-1
Source: https://github.com/black-yt/skills/tree/main/lab-cluster-1
Command: npx skills add https://github.com/black-yt/skills --skill lab-cluster-1

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps you work safely and efficiently on the lab cluster by standardizing login, file editing, job submission, service deployment, networking, and environment boundaries. It reduces mistakes that can damage shared resources, leak secrets, or waste GPU and CPU quota.

Core Features & Use Cases

  • Cluster Access and Editing: Connect to the development host, keep a persistent SSH session, and edit remote files with validated patch or copy workflows.
  • Job and Worker Operations: Launch temporary workers, submit CPU or GPU jobs, inspect logs and events, and clean up resources correctly.
  • Networking and Services: Configure proxies for CPU tasks, handle private-network access, deploy internal services, and connect to them safely from local or remote machines.
  • Storage and Model Handling: Use the correct shared paths for projects, large files, caches, and Hugging Face model directories, including cache-to-model conversion.

Quick Start

Ask me to prepare the safest cluster workflow for your task, and I will help you choose the right SSH, worker, job, storage, and network commands.

Frequently Asked Questions about lab-cluster-1

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

FAQPage Schema
How do I submit GPU jobs on a lab cluster safely?

To submit GPU jobs safely on a lab cluster, this Skill standardizes rjob submissions and rlaunch worker operations with enforced resource rules. It prevents wasted quota and ensures correct GPU allocation.

What's the best way to manage SSH access and remote editing on shared development hosts?

Managing SSH access on shared development hosts involves keeping persistent sessions and using validated patch or copy workflows for remote editing. This prevents accidental damage to shared resources during file modifications.

How do I configure proxies for CPU tasks and private-network services?

Configuring proxies for CPU tasks requires specific proxy handling and namespace selection rules. This Skill manages private-network access and deploys internal services safely for remote connections.

Can I use Hugging Face model directories and shared storage paths on a lab cluster?

Yes, you can use Hugging Face model directories on a lab cluster by applying correct shared storage paths for projects, large files, and caches. This includes cache-to-model conversion and enforces cache discipline.

Why do my rlaunch workers or rjob submissions fail to clean up resources correctly?

Your rlaunch workers or rjob submissions may fail due to missing command safety enforcement. This Skill standardizes cleanup workflows, log inspection, and resource rules to prevent wasted CPU and GPU quota.