Automate ML experiment design, execution, and archival via a 6-stage workflow.
npx skills add https://github.com/schmidtkk/skills --skill auto-research-schmidtkk
Please help me install this Agent Skill. Skill: auto-research Source: https://github.com/schmidtkk/skills/tree/main/auto-research Command: npx skills add https://github.com/schmidtkk/skills --skill auto-research-schmidtkk
Auto Research provides a disciplined, reproducible framework to design ML experiments, run them in a controlled loop, and capture lessons.
Follow the 6-stage loop to design an ML experiment, approve the plan, run, analyze, and archive lessons.
High-intent search queries and answers about installing and using this skill.
To structure ML experiments for reproducibility, use a 6-stage workflow enforcing INTAKE, PLAN, RUN, REFINE, EVAL, and ARCHIVE phases with deterministic artifacts like experiment.py and results.json for traceable archival.
Automate ablation studies by running controlled iterations through the 6-stage research loop, which automatically captures post-run analysis in analysis.md and archives lessons in lessons.md for robust knowledge retention.
No specific dependencies are required to run structured AI experimentation. The framework operates independently to enforce deterministic code execution via experiment.py and manages artifacts like plan.yaml without external tool requirements.
The best way to archive ML research artifacts is through the ARCHIVE stage, which systematically stores task.md, plan.yaml, experiment.py, results.json, analysis.md, and lessons.md to ensure complete traceability across experiments.
pluginagentmarketplaceBuilds configurable ML pipelines for preprocessing, feature engineering, model evaluation, and validation.
danny0926Analyze backtest results with AB comparisons and bias screening.
OKHP3Convert AI-driven assets and ideas into standardized GitHub repositories.
lengoctuong2005Automate logging and tracking of machine learning experiments with wandb.
