skills.rest
jiahao-shao1
jiahao-shao1COMMUNITY
Explore all skills by this author →
Website𝕏@jiahaoshao1GitHub

remote-cluster-agent

Execute remote commands on GPU clusters via a persistent SSH agent.

9|1|Updated Mar 19, 2026
Search Tags:#automation#mcp#gpu#ssh#remote#cluster#mutagen
One-click install
npx skills add https://github.com/jiahao-shao1/remote-cluster-agent --skill remote-cluster-agent
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: remote-cluster-agent
Source: https://github.com/jiahao-shao1/remote-cluster-agent/tree/main
Command: npx skills add https://github.com/jiahao-shao1/remote-cluster-agent --skill remote-cluster-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enables fast remote command execution on GPU clusters by using a persistent SSH connection and a cluster-side agent, reducing latency and manual setup.

Core Features & Use Cases

  • Low-latency remote execution: Run commands on remote GPUs with ~0.1s latency via a persistent SSH agent and agent-mode MCP server.
  • Mutagen-based code synchronization: Real-time local-to-remote code sync to support iterative development workflows.
  • Multi-node routing: Single MCP server can route commands to multiple cluster nodes using the node parameter.

Quick Start

Install the skill, complete the interactive first-time setup to configure your cluster, deploy the MCP server, then start using remote_bash to run commands on the GPU cluster.

Frequently Asked Questions about remote-cluster-agent

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

FAQPage Schema
How do I run commands on a remote GPU cluster with low latency over SSH?▼

You can achieve ~0.1s latency for remote GPU command execution by using a persistent SSH agent and an agent-mode MCP server, which removes the overhead of establishing a new connection for every command.

How do I sync code to a remote GPU cluster for iterative development?▼

Mutagen-based code synchronization provides real-time local-to-remote sync to support iterative development workflows on GPU clusters. This ensures your remote environment matches local changes instantly without manual file transfers.

Can I route commands to multiple nodes in a GPU cluster through a single MCP server?▼

Yes, a single MCP server can route commands to multiple cluster nodes using the node parameter, allowing centralized multi-node remote execution through one interface without separate connections.

What do I need to set up remote execution on a GPU cluster?▼

Setting up remote GPU cluster execution requires deploying a cluster-side agent, an MCP server with node routing, and Mutagen for real-time sync, plus an interactive first-time setup to configure your environment.

Does remote GPU cluster execution work without maintaining a persistent SSH connection?▼

A persistent SSH connection is required to achieve low-latency remote execution on GPU clusters. Without it, standard SSH overhead would reintroduce delays for every per-command execution cycle.

Related Skills

Explore More Software Engineering Skills→
umich-foreseerumich-foreseer
16

connect

Automate cross-cluster SSH tunnel setup between Great Lakes and Lighthouse.

Official
Advanced
primatrixprimatrix

exec-remote

Execute Python scripts on remote GPU or TPU clusters via SkyPilot.

Official
Advanced
gqf2008gqf2008

lambda-labs-gpu-cloud

Provision on-demand Lambda Labs GPU instances with SSH and persistent filesystems.

Community
Intermediate
AirmomoAirmomo

ssh-essentials

Manage SSH keys, tunnels, and file transfers with OpenSSH tools.

Community
Intermediate