run-experiment

Deploy machine learning experiments to local or remote GPU servers via SSH.

Updated May 22, 2026
One-click install
npx skills add https://github.com/Leo1349/autoresearch --skill run-experiment-leo1349
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: run-experiment
Source: https://github.com/Leo1349/autoresearch/tree/main/skills/run-experiment
Command: npx skills add https://github.com/Leo1349/autoresearch --skill run-experiment-leo1349

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill addresses the complexity of deploying machine learning experiments on local or remote GPU servers, streamlining the process and reducing potential errors.

Core Features & Use Cases

  • Experiment Deployment: Automates the deployment of experiments to local or remote GPU servers.
  • Environment Detection: Identifies the experiment environment (local GPU, remote server).
  • Pre-flight Check: Verifies GPU availability on the target machine.
  • Code Sync: Syncs code to the remote server, if necessary.
  • W&B Integration: Optionally integrates Weights & Biases for experiment tracking.
  • Experiment Launch: Executes the experiment on the target machine.
  • Notification: Sends a notification upon experiment completion.
  • Use Case: Ideal for researchers and data scientists looking to deploy experiments without manually configuring server environments.

Quick Start

Use the run-experiment skill to deploy an experiment to a remote server.

Frequently Asked Questions about run-experiment

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

FAQPage Schema
How do I deploy machine learning experiments to a remote GPU server?

The Skill automates deployment of machine learning experiments to remote GPU servers by detecting the environment, syncing code, running pre-flight GPU checks, and launching execution via SSH access. This removes the need for manual server setup.

Do I need SSH access to run ML experiments on remote GPU servers?

Yes, SSH access is required to run ML experiments on remote GPU servers. The Skill uses SSH to connect to the target machine, verify GPU availability, synchronize code, and execute the experiment remotely.

Can I use Weights and Biases for experiment tracking when deploying to GPU servers?

Yes, Weights and Biases integration is optionally supported for experiment tracking when deploying to GPU servers. This enables automated monitoring of runs and metrics after the experiment launches on the target machine.

What is the best way to automate code sync and pre-flight checks for ML experiment deployment?

Automating code sync and pre-flight checks for ML experiment deployment is handled by detecting the environment and verifying GPU availability on the target machine before execution. This ensures the remote server is ready before launching the experiment.

Does this deployment automation work with local GPU environments or only remote servers?

This deployment automation works with both local GPU environments and remote servers. It detects the experiment environment, identifies available GPUs, and synchronizes code only when a remote target is specified.

How do I get notified upon experiment completion on a remote GPU server?

You receive a notification upon experiment completion on a remote GPU server through the deployment automation's built-in alert feature. After launching the experiment and optionally tracking it via W&B, the Skill sends a completion alert.