experiment-agent

Execute and monitor code experiments and human studies with stall detection.

170|11|Updated Apr 9, 2026

Search Tags:#monitoring #reproducibility #experiment #code-runner #ethics-checks #ars-integration #material-passport

Ready to run

npx skills add https://github.com/Imbad0202/experiment-agent --skill experiment-agent

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: experiment-agent
Source: https://github.com/Imbad0202/experiment-agent/tree/main
Command: npx skills add https://github.com/Imbad0202/experiment-agent --skill experiment-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Executes and monitors experiments in academic research to ensure reliable execution, transparent interpretation of results, and reproducible workflows.

Core Features & Use Cases

Run code experiments (ML training, statistical analysis, ETL, simulation) with real-time monitoring for stalls or crashes.
Manage human studies (protocol planning, ethics checks, data collection progress).
Interpret statistics (p-values, effect sizes, CIs) and verify reproducibility; produce ARS-compatible outputs.
Operate in four modes: run, manage, validate, plan; with integration to Material Passport outputs.

Quick Start

Start Claude Code and run your desired experiment command, for example: Run my training script: python train.py.

Frequently Asked Questions about experiment-agent

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

FAQPage Schema

How do I monitor code-based experiments for stalls or crashes in real-time?▼

To monitor experiments, run your code-based scripts such as ML training, statistical analysis, ETL, or simulation through the system to receive real-time stall detection and crash alerts during execution.

How do I ensure reproducibility when running statistical analysis and ML training scripts?▼

To ensure reproducibility, the validate mode interprets statistics like p-values, effect sizes, and confidence intervals, verifying that code-based experiment workflows produce consistent, transparent results.

Can I manage human studies and ethics checks alongside code-run experiments?▼

Yes, the manage mode supports human studies by handling protocol planning, verifying ethics checks, and tracking data collection progress for surveys, field work, and interviews alongside code experiments.

Does experiment monitoring work with ARS integration and Material Passport outputs?▼

Yes, experiment execution and validation optionally integrate with ARS by generating Material Passport outputs, ensuring your code-run and human study results are compatible and reproducible across platforms.

What are the limitations of automating reproducibility checks for academic research experiments?▼

Reproducibility validation is limited to interpreting statistics like p-values, effect sizes, and confidence intervals from code-run experiments, requiring properly structured scripts to generate ARS-compatible Material Passport outputs.