What problem does it solve? Running reproducible data analysis requires writing code, inspecting unfamiliar data files, managing execution environments, and documenting methodology. This Skill handles the full cycle: it inspects Excel/CSV schemas, executes Python or R analysis code in a sandbox, and exports structured results with complete methodology documentation. ## Core Features & Use Cases - Data Inspection: Automatically loads Excel and CSV files into DuckDB to report sheet names, column types, row counts, non-null counts, and sample rows before any analysis begins. - Python/R Code Execution: Runs inline code snippets or script files in a subprocess with automatic data-file loading, package auto-installation, and a 300-second timeout. - Structured Result Export: Captures a DataFrame assigned to result and exports it as CSV, JSON, or Markdown with proper headers and rows. - Use Case: Given a sales dataset in Excel, inspect its schema, run a Pearson correlation analysis between two variables in Python, and export the aggregated results as a JSON file alongside documented methodology and data traceability. ## Quick Start Use the code-engineer skill to inspect the columns in sales_2024.xlsx and then run a Python analysis computing total sales by region, exporting the result to results.csv.
