data-aggregation-and-reporting

name	data-aggregation-and-reporting
description	Aggregate data by multiple dimensions (time, product, region, channel) with configurable metrics, pivot tables, and generate KPI summary reports with top products, trends, and cross-dimensional breakdowns.
version	1.0.0

Data Aggregation & Reporting

Skill Summary

Performs group-by aggregations across multiple dimensions and generates comprehensive KPI summary reports. Supports flat aggregation, pivot table generation, rolling window aggregation, and pre-built sales reporting templates (daily/weekly/monthly trends, top products, regional/channel breakdowns, category analysis).

Merges functionality from: data_aggregator, summary_report_generator.

Aggregation Modes

• Flat: Standard group-by with configurable metrics (sum, mean, count, min, max, median, std)
• Pivot: Cross-tabulation with row/column dimensions
• Rolling: Time-based rolling window aggregation

Report Sections (for sales data)

1. Overview KPIs: Total revenue, total orders, average order value, total units
2. Time trends: Daily, weekly, monthly revenue and quantity
3. Top products: Top N by revenue and by units sold
4. Regional breakdown: Revenue and orders by region
5. Channel performance: Online vs retail vs wholesale
6. Category analysis: Revenue by product category
7. Cross-dimensional: Region × Category revenue matrix

Inputs

• --input / -i: Path to input data (CSV or Parquet) (required)
• --group-by / -g: Grouping columns (comma-separated)
• --metrics: Metric specs like revenue:sum,quantity:mean,order_id:count
• --mode: flat, pivot, rolling, or report (default: flat)
• --pivot-columns: Column for pivot columns
• --pivot-values: Value column for pivot
• --window: Window size for rolling mode
• --catalog / -c: Product catalog JSON (for category info in report mode)
• --top-n: Number of top items in report (default: 10)
• --time-granularity: daily, weekly, monthly (default: all)
• --output / -o: Output directory
• --sort-by: Sort by column (descending)

Output

• Flat/Pivot/Rolling mode: Aggregated dataset (CSV or Parquet)
• Report mode:
  • summary_report.json: KPI summary with all sections
  • agg_daily.csv, agg_weekly.csv, agg_monthly.csv: Time aggregations
  • agg_by_product.csv: Per-product summary
  • agg_by_region.csv: Regional breakdown
  • agg_by_channel.csv: Channel performance
  • agg_region_category.csv: Region × Category pivot

Implementation

• Command: python3 ./skills/data_aggregation_and_reporting/aggregate.py -i <file> -o <output_dir>
• Dependencies: pandas, numpy