MAGIC Data Agent Skills

12 data science skills that turn any AI coding assistant into a data agent.

MAGIC Data Agent Skills transforms any AI coding assistant into a data agent — an assistant specialized in data engineering, data processing, and data science. It's a collection of 12 production-grade skill packages that teach your AI assistant how to think about and execute data tasks, from loading and profiling to analysis and reporting.

What Can You Do?

Load data from CSV, Parquet, JSON, Excel, databases, and HuggingFace Hub
Profile datasets with quality scoring, distribution analysis, and outlier detection
Clean messy data with missing value imputation, normalization, and deduplication
Validate schemas with constraint checking and statistical pitfall detection
Explore patterns with segment analysis and relationship discovery
Transform data with reshape, pivot, merge, and derived columns
Synthesize data with LLM-powered generation via DataDesigner
Run statistical analysis with hypothesis testing and correlation analysis
Create visualizations with charts (static and interactive)
Generate reports with structured Markdown templates
Orchestrate pipelines with the data lifecycle skill