access-gpu-sandbox

Provision SSH user accounts and connect to the McGill-NLP GPU sandbox.

Updated Apr 24, 2026
One-click install
npx skills add https://github.com/gazaille4mila/agent-review-competition --skill access-gpu-sandbox
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: access-gpu-sandbox
Source: https://github.com/gazaille4mila/agent-review-competition/tree/main/.claude/skills/access-gpu-sandbox
Command: npx skills add https://github.com/gazaille4mila/agent-review-competition --skill access-gpu-sandbox

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill enables users to securely connect to the McGill-NLP shared GPU sandbox through SSH, provisioning individual user accounts for isolated access.

Core Features & Use Cases

  • User provisioning: Automates creation of dedicated SSH user accounts in the GPU sandbox environment.
  • Secure connection: Establishes SSH sessions for remote GPU access, ensuring data safety.
  • Use Case: Researchers or developers needing isolated GPU access can quickly provision a user and connect to run their workloads on the shared hardware.

Quick Start

Provision a new user and connect to the GPU sandbox by generating SSH keys, registering the user via the API, and then SSH-ing with the assigned credentials.

Frequently Asked Questions about access-gpu-sandbox

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

FAQPage Schema
How do I securely access shared GPU resources via SSH?

Securely access shared GPU resources by provisioning a dedicated user account and managing SSH keys, which establishes an isolated SSH session for remote GPU development workflows.

What is GPU sandbox user provisioning and how does it work?

GPU sandbox user provisioning automates the creation of dedicated SSH user accounts, generating and registering SSH keys via an API to ensure isolated access to shared hardware.

Can I use SSH to run workloads on a shared GPU sandbox?

Yes, you can use SSH to run workloads on a shared GPU sandbox by provisioning an isolated user account and connecting with the assigned credentials.

How do I generate and register SSH keys for remote GPU access?

Generate SSH keys locally, then register the user and public key via the sandbox API to receive assigned credentials for establishing a secure SSH connection to the GPU.

Do I need an isolated user account to connect to a shared GPU environment?

Yes, an isolated user account is required to connect to the shared GPU environment, ensuring data safety and preventing interference from other researchers on the same hardware.