Imbad0202Imbad0202Communityยท1 Agent Skills Included

experiment-agent

Run, monitor, and validate academic research experiments

Executes code experiments, manages human studies, interprets statistics, and verifies reproducibility for academic research. Eliminates manual process monitoring, missed statistical fallacies, and lost study progress across long research timelines. Keeps experiments safe with user-controlled execution, IRB ethics checks, and structured reports ready for paper writing.
npx skills add Imbad0202/experiment-agent --all -g -y
Available:

Tells the agent how to route requests between run, manage, validate, and plan modes, and enforces safety rules like advisory-only anomaly detection and user-approved command execution.

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install experiment-agent?โ–ผ

Run `npx skills add Imbad0202/experiment-agent --all -g -y` in your terminal to install the skill globally.

How to run and monitor a code experiment?โ–ผ

Use run mode with a command like "Run my training script: python train.py" and it executes the script while watching for stalls, crashes, and timeouts in real time.

Can it check my statistics for errors?โ–ผ

Yes. Validate mode interprets p-values, effect sizes, and confidence intervals, and scans for 11 common statistical fallacies like Simpson's Paradox and survivorship bias.

Does it help manage human studies and surveys?โ–ผ

Yes. Manage mode plans study protocols, checks IRB ethics requirements, tracks participant progress, and saves state so long studies survive session restarts.

Does experiment-agent work with Claude Code?โ–ผ

Yes. It is built as a standard SKILL.md skill and runs directly in Claude Code and other compatible runtimes.

Related Repositories in Education & Research

View All in Education & Researchโ†’