monitor-experiment

Monitor remote experiments via SSH and collect results from logs.

Updated Mar 15, 2026
One-click install
npx skills add https://github.com/MACJONSW/AutoResearchIDEA --skill monitor-experiment-macjonsw
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: monitor-experiment
Source: https://github.com/MACJONSW/AutoResearchIDEA/tree/main/skills/monitor-experiment
Command: npx skills add https://github.com/MACJONSW/AutoResearchIDEA --skill monitor-experiment-macjonsw

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill helps users monitor the progress and collect the results of running experiments, providing clear insights into their status and outcomes.

Core Features & Use Cases

  • Real-time Monitoring: Check the status of running experiments via SSH and screen sessions.
  • Result Collection: Automatically fetch and parse output logs and JSON result files.
  • Interpretation & Reporting: Summarize findings in a comparison table and flag unexpected results.
  • Use Case: After initiating a long-running experiment, use this Skill to check if it's completed, gather the key metrics, and see how it performed against the baseline.

Quick Start

Monitor the experiment named 'my-research-run' on the server.

Frequently Asked Questions about monitor-experiment

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

FAQPage Schema
How do I track the progress of a running experiment on a remote server?

Track running experiment progress on a remote server by using SSH to connect and check the status of active screen sessions. This allows you to verify whether your long-running processes are still active or have completed.

How do I collect and parse results from remote experiment logs?

Collect and parse remote experiment results by fetching output logs and JSON result files via SSH. The skill automatically extracts these files from the server and reads the key metrics generated by your bash commands.

How can I compare experiment metrics against a baseline and flag anomalies?

Compare experiment metrics against a baseline by parsing the collected JSON outputs and log files. The skill summarizes findings in a comparison table and automatically flags unexpected results or anomalies.

Do I need SSH access to monitor experiments with this tool?

Yes, SSH access is required to monitor experiments because the tool connects to remote servers via SSH. It relies on bash commands and screen sessions on the remote host to check status and collect output files.

Can I receive notifications when a remote experiment is completed?

You can receive notifications when a remote experiment is completed through optional Feishu notifications. This alerts you the moment the monitoring script detects that the screen session has finished running.

What is the best way to monitor long-running bash experiments remotely?

The best way to monitor long-running bash experiments is to use a tool that checks screen sessions via SSH, collects JSON outputs, compares metrics against baselines, and sends a Feishu notification upon completion.