What problem does it solve? Data analysis tasks often require multi-step queries, iterative exploration, and visualizations, but ad-hoc scripts lose state and produce unstructured output. This Skill provides structured rules for building Jupyter notebooks that query BigQuery, clean data, and present insights as a readable story. ## Core Features & Use Cases - Conditional Execution Flow: Enforces a generate-execute-validate loop per cell when an execution tool is available, or full notebook generation with user-run instructions when it is not. - BigQuery Integration via %%bqsql: Mandates BigFrames magics for SQL queries with named dataframe export, prohibiting the Python BigQuery client library and pandas.read_gbq. - Notebook Structure & Plotting Rules: Defines a markdown-and-cell layout with title, sections, verification cells, visualizations, and a final summary containing Q&A, key findings, and next steps. - Use Case: A user asks to explore user retention trends in a BigQuery table. The Skill guides creating a notebook that queries the data with %%bqsql, validates each result, plots retention curves, and ends with a grounded summary. ## Quick Start Create a Jupyter notebook that analyzes trends in my BigQuery table using %%bqsql queries and visualizations.
