Data Scientist
Design experiments, profile datasets, build models, and audit them for bias before shipping
The Data Scientist plugin helps you move from a dataset to a defensible production model. Every command works standalone with your input — describe the experiment, dataset, or model and get a structured draft ready for your review.
Use /design-experiment to plan an A/B test or experiment with sample size, power analysis, and recommended duration. Use /analyze-dataset to ingest a CSV or Parquet description, generate a data-quality report, flag outliers and biases, and suggest transforms. Use /build-model to recommend a model architecture, generate a scikit-learn or XGBoost template, and document assumptions. Use /eval-model to compute cross-validation, generate a confusion matrix, and audit feature importance for bias.
The plugin's skills activate automatically when relevant: Dataset Profiling auto-scans for missing values, outliers, class imbalance, correlation issues, and schema drift, plus Model Card Generation auto-generates Model Cards covering intended use, limitations, and training data provenance per the HuggingFace and Mitchell et al. standard.
Commands
/design-experimentPlan an A/B test or experiment with sample size, power analysis, and recommended duration
/analyze-datasetGenerate a data quality report from a dataset description, flag outliers and biases, suggest transforms
/build-modelRecommend model architecture, generate scikit-learn or XGBoost template, document assumptions
/eval-modelCompute cross-validation, generate confusion matrix, audit feature importance for bias
Skills
Skills activate automatically when Claude detects you're working on relevant tasks — no slash command needed.
Dataset Profiling
Auto-scans for missing values, outliers, class imbalance, correlation issues, and schema drift
Model Card Generation
Auto-generates Model Cards (intended use, limitations, training data provenance) per HuggingFace and Mitchell et al. standard
Usage examples
/design-experiment Plan A/B test on checkout flow, baseline 4.2% conversion, MDE 10% relative, 12K daily visitors — sample size and duration
/analyze-dataset 240K-row churn dataset, 14 numeric / 22 categorical / 2 text — generate quality report and preprocessing recommendations
/build-model Predict 30-day churn (binary, ~12% positive) on tabular data — recommend architecture and generate scikit-learn pipeline
/eval-model Evaluate XGBoost churn classifier with 5-fold CV, confusion matrix at threshold 0.3, and bias audit on region and age band
Install this plugin
/plugin install data-scientist@alexclowe/awesome-claude-cowork-pluginsRun this command in Claude Cowork to install. Requires a paid Claude plan (Pro, Max, Team, or Enterprise).
All content generated by this plugin is for drafting and informational purposes only. The data scientist is responsible for reviewing, verifying, and customizing all outputs before professional use. This plugin does not constitute professional advice.
More plugins
Pharmacist
Review drug interactions, check dosing protocols, and draft patient counseling notes
Physical Therapist
Draft SOAP notes, build exercise programs, and write insurance appeal letters
Dental Hygienist
Document periodontal findings, write insurance narratives, and create patient education materials
Chiropractor
Chart SOAP notes, draft medical necessity letters, and generate practice marketing content
Recommended tools for your workflow
AI Agent Store
Free to browseDirectory of AI agents and tools across categories — discover, compare, and adopt agents built for your workflow.
Some links on this page are affiliate links. We may earn a commission if you purchase — at no extra cost to you. We only recommend tools we believe in for your profession.