- Home
- Skills
- Sammcj
- Agentic Coding
- Storytelling With Data
storytelling-with-data_skill
- Python
110
GitHub Stars
2
Bundled Files
3 weeks ago
Catalog Refreshed
1 month ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstart where the catalogue uses aiagentskills.
npx veilstart add skill sammcj/agentic-coding --skill storytelling-with-data- README.md137 B
- SKILL.md12.0 KB
Overview
This skill applies the Storytelling with Data (SWD) six-lesson framework to create, review, and improve data visualisations and data-driven communications. It focuses on clarifying the single takeaway, choosing the right visual, decluttering, guiding attention, applying good design, and structuring a narrative arc. Use it for charts, dashboards, infographics, slide decks, reports, or data-driven web pages. The guidance is format-agnostic and pairs with format-specific tools for implementation.
How this skill works
I start by clarifying context: audience, the one action or takeaway, and the delivery medium. Then I pick the simplest effective visual for each message, remove unnecessary elements, and apply a grey-plus-one-accent colour strategy to direct attention. I ensure each panel or slide has a declarative title, readable annotations, and a clear narrative flow (setup, tension, resolution). For reviews or makeovers I provide concrete before/after recommendations and a short checklist of fixes.
When to use it
- Creating new charts, dashboards, infographics, slide decks, or data-driven pages.
- Reviewing existing visuals to improve clarity, impact, or accessibility.
- Performing chart makeovers to simplify, declutter, and highlight the key message.
- Choosing the right chart type and visual strategy for a specific audience and goal.
- Advising how to structure a narrative or sequence of panels/slides for persuasion.
Best practices
- Start with audience + one-sentence "big idea" before touching a tool; storyboard first.
- Match visual type to message (bars for comparison, lines for trends, text for single KPIs).
- Eliminate clutter: remove unnecessary borders, gridlines, markers, and redundant labels.
- Use grey for defaults and one accent colour to highlight what matters; label directly where possible.
- Give every chart a declarative action title and add concise annotations to explain the "so what".
- Design for accessibility and progressive disclosure based on whether a presenter will narrate the visuals.
Example use cases
- Turn a dense monthly sales dashboard into an executive-ready overview with drill-down panels.
- Make over a multi-line 'spaghetti' chart into small multiples or focused comparisons.
- Convert exploratory analysis into an explanatory slide deck with a clear beginning–middle–end narrative.
- Design an infographic that reads top-to-bottom with clear section headers and a single accent colour.
- Review an emailed report and add annotations and declarative titles so it stands alone without a presenter.
FAQ
Provide an explanatory summary up front and include supporting detail on demand (appendices, drill-downs, or separate tables) so the main view stays focused.
When is a table better than a chart?
Use a table for precise lookups or many sortable rows; use a chart when the pattern, trend, or comparison is the key insight.