FastReports

FastReports eliminates the repetitive hours spent on data preparation and exploratory analysis. Drop in a CSV, Excel, JSON, or Parquet file and an 8-phase pipeline runs automatically: it ingests and profiles the data, scores data quality, generates and applies cleaning strategies, runs statistical analysis and EDA, produces interactive Plotly visualizations, and renders a self-contained HTML report with embedded charts — all in under a second. IBM Bob is embedded throughout the pipeline, not bolted on as an afterthought. Bob assists with architecture design in Plan Mode, generates module implementations in Code Mode, recommends cleaning strategies per column type, interprets statistical findings, and selects chart types based on data characteristics. Every interaction is logged to a session file for full traceability. The project ships in three deployment modes: a single CLI command for local analysis, a FastAPI + DuckDB REST API for programmatic access, and an interactive Preact dashboard with real-time column filtering, arbitrary DuckDB SQL queries, and dynamic chart generation. Built with 9,269 lines of Python across 33 modules and 942 lines of frontend code.