Manage remote GPU servers via tm CLI for commands, jobs, and monitoring.
npx skills add https://github.com/justcyl/my-skills --skill tm
Please help me install this Agent Skill. Skill: tm Source: https://github.com/justcyl/my-skills/tree/main/.skills/archived-skills/tm Command: npx skills add https://github.com/justcyl/my-skills --skill tm
Remote GPU server management often requires ad-hoc SSH sessions, inconsistent environments, and manual coordination across hosts. tm provides a consistent CLI workflow to execute commands, run GPU-bound training/evaluation jobs, sync project code, and monitor GPU availability.
Use tm to run a quick remote command on a server to verify connectivity.
High-intent search queries and answers about installing and using this skill.
You can automate remote GPU server management and execute commands without active SSH sessions using the tm CLI. It supports stateless remote control to run short executions via tm exec and manage long-running jobs in the background with tm run and tm wait.
The best way to monitor GPU availability across multiple remote hosts is using tm gpu for status reporting. It provides a consistent CLI workflow to check GPU resources and verify availability as part of your reproducible remote server workflows.
You can sync project code to a remote server for training and evaluation jobs using the tm CLI. It automates code synchronization across hosts to ensure consistent environments and reproducible workflows before launching GPU-bound tasks.
Yes, you can orchestrate parallel experiments on remote GPU servers from a local machine using the tm CLI. It enables launching long-running jobs and managing parallel background workers to coordinate GPU-bound training and evaluation across multiple hosts.
Yes, the tm CLI handles long-running training jobs even if your local terminal disconnects. By using tm run to launch tasks and tm wait to manage them, the workflow operates statelessly over SSH to ensure background jobs continue running on the remote server.
You should use tm exec for short executions to quickly verify connectivity or run brief commands, whereas tm run is designed for launching long-running jobs and background workers that require lifecycle management via tm wait on the remote GPU server.
djcyphersExecute remote commands on Windows servers via WinRM, PowerShell remoting, and SSH.
NVIDIA-TAOAutomate remote TAO GPU job submission on SLURM clusters via SSH.
the-open-agentAutomate GitHub repository creation, authentication, and code pushes.
qcmuuOrchestrate ML training and batch jobs across multiple cloud providers with SkyPilot.
