Research: data-wizard

Evidence only

Research cache for catalog enrichment. Not authoritative for install or trust decisions.

Back to catalog page

Quick Answer

Problem: Analyze data and guide ML: EDA, model selection, feature engineering, stats, visualization, MLOps. Use for data work. NOT for ETL, database design (database-architect), or frontend viz code.

Stack / assumptions: portable skill scripts under scripts/; on-demand references; eval fixtures

Comparable alternative: database-architect for etl, database design

Repo summary:

Full-stack data science and ML engineering — from exploratory data analysis through model deployment strategy. Adapts approach based on complexity classification.

> Grounded in repository skills/data-wizard/SKILL.md; treat as evidence, not authority.

Viz Pipeline (Wave 9)

Sub-modes: viz plan → viz render → optional viz dashboard for shareable EDA HTML.

Mode	Script	Output
Viz Plan	scripts/viz-planner.py	JSON chart plan (goal_category, encodings, column mappings)
Viz Render	scripts/viz-renderer.py	PNG or HTML chart files from plan JSON
Viz Dashboard	scripts/dashboard-builder.py	Composable HTML dashboard from profiler JSON

Flow: Profile data with data-profiler.py first; use base viz for recommendations only. When the user wants artifacts, run plan then render; pass --viz-plan and --render-result into the dashboard builder when embedding charts.

References: references/visualization.md (plan/render), references/dashboard-design.md (dashboard schema).