storytelling-viz

What problem does it solve?

This Skill eliminates the guesswork and manual effort of turning raw datasets into clear, stakeholder-ready interactive visualizations. It ensures your chart has a single defensible takeaway, stays honest to the underlying data, and includes all necessary context and caveats, without requiring you to handle website publishing or dashboard development.

Core Features & Use Cases

• Dataset Validation & Chart Recommendation: Automatically inspects datasets to verify sources, check comparability, and recommend one default chart plus one fallback tailored to your data's story.
• Stakeholder-Ready Artifacts: Produces self-contained index.html and preview.html files, plus 2-3 insight bullets and explicit source citations, ready for local review or embedding.
• Built-In Quality Guardrails: Enforces non-negotiable honesty, intuition, and concision tests, requires user confirmation before building when multiple chart directions are plausible, and includes rendered visual QA to catch clipping, overlap, or spacing issues.
• Use Cases: Ideal for analyst-to-stakeholder charts, one-off story visualizations, visual explainers for reports or presentations, and example galleries for chart-generation workflows. It works portably in both Codex and Claude Code environments.

Quick Start

Use the storytelling-viz skill to inspect my attached dataset, recommend the best chart and one fallback, wait for my confirmation, then build a polished Plotly visualization with index.html and preview.html.