skills.rest
jandan138
jandan138COMMUNITY
Explore all skills by this author →
GitHub

monitor-experiment

Monitor running experiments via SSH and Vast.ai, collecting screen session progress and JSON results.

Updated Apr 26, 2026
Search Tags:#metrics#ssh#experiments#monitor#wandb#screen
One-click install
npx skills add https://github.com/jandan138/Auto-claude-code-research-in-sleep --skill monitor-experiment-jandan138
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: monitor-experiment
Source: https://github.com/jandan138/Auto-claude-code-research-in-sleep/tree/main/skills/monitor-experiment
Command: npx skills add https://github.com/jandan138/Auto-claude-code-research-in-sleep --skill monitor-experiment-jandan138

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps you monitor long-running experiments, collect progress, and unify results from multiple sessions and runs.

Core Features & Use Cases

  • Monitor running experiments via SSH and Vast.ai instances.
  • Collect the last N lines of each screen session to assemble live status.
  • Check for JSON result files and fetch final metrics.
  • Optional: pull Weights & Biases metrics when wandb is true in CLAUDE.md for deeper insights.

Quick Start

Monitor the running experiments on my SSH server and report the latest progress and results.

Frequently Asked Questions about monitor-experiment

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

FAQPage Schema
How do I monitor running machine learning experiments via SSH?▼

Monitor running experiments via SSH by connecting to remote servers to inspect active screen sessions and retrieve the latest output lines. It collects live progress updates and final metrics from long-running machine learning runs across SSH and Vast.ai instances.

How do I collect JSON result files from distributed runs on Vast.ai?▼

Collect JSON result files from distributed runs on Vast.ai by discovering and fetching them directly from the remote instance. This skill checks for JSON result files to retrieve final metrics after your long-running experiments complete.

Can I pull Weights and Biases metrics for my remote experiments?▼

Yes, you can pull Weights and Biases metrics for remote experiments when wandb is set to true in your CLAUDE.md configuration. This optional feature retrieves deeper insights and metrics from your configured wandb runs.

Does this experiment monitor work with screen sessions on remote servers?▼

Yes, this experiment monitor works with screen sessions on remote servers by collecting the last N lines of each session to assemble a live status report. It applies to long-running machine learning experiments across SSH and Vast.ai environments.

What is the best way to track progress across multiple long-running experiments?▼

The best way to track progress across multiple long-running experiments is to unify results from multiple sessions and runs. This skill inspects screen sessions, retrieves JSON result files, and optionally pulls wandb metrics to assemble a comprehensive status overview.

Related Skills

Explore More Data & Analytics Skills→
JuanMS20JuanMS20

github-auth

Configure GitHub authentication for Hermes Agent using git, gh CLI, or SSH credentials.

Community
Intermediate
thisismynewfmail-uithisismynewfmail-ui

weights-and-biases

Log machine learning experiment metrics and visualize training progress with W&B.

Community
Intermediate
raja21068raja21068
2

run-experiment

Deploy and run machine learning experiments across local GPUs, SSH servers, Vast.ai, and Modal.

Community
Advanced
satsuki-64satsuki-64
2

monitor-experiment

Monitor experiment progress via SSH and parse JSON results from remote servers.

Community
Advanced