data-storytelling
Data Storytelling for Slides: Charts, graphs, data visualization, and narrative techniques for presenting data in presentations
Profile →npx skills add cosmicstack-labs/mercury-agent-skills --skill data-storytelling --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Turn raw data into compelling visual narratives that audiences understand, remember, and act on.
How it works
The skill instructs the agent to:
- Emphasize that data needs a narrative, not just numbers, and to show relationships like cause/effect or target vs. actual.
- Promote maximizing the data-ink ratio by removing non-essential chart elements and focusing on mappings between data points and insights.
- Prioritize quick audience comprehension, ensuring the key takeaway is identifiable within 5 seconds using pre-attentive attributes (color, size, position).
- Use a maturity framework to select appropriate chart types and narrative integration (ranging from raw data to immersive, multi-chart storytelling), with target levels specified for internal dashboards vs investor decks.
- Provide a detailed Chart Selection Guide to choose chart types (line, bar, histogram, scatter, etc.) based on data relationships and a decision flowchart for charting options.
- Include guidance on when not to use pie charts, and when to apply specific chart types for time, composition, distribution, and flow.
- Offer actionable storytelling templates, including the Three-Act Data Story structure (Context, Insight, Implication) and a data-narrative Python template for slide formatting.
- Demonstrate data-ink improvements with before/after examples and Python code snippets for creating cleaner charts.
- Outline pre-attentive attributes (color, intensity, size, position, orientation, shape, motion) and provide practical highlighting techniques via Python functions that emphasize key data points.
- Propose a Dashboard Slide Design section illustrating executive dashboards, KPI patterns, and waterfall KPI layouts.
- Present Annotated Charts guidelines, five annotation types (Callout, Trend line, Threshold, Data label, Comparative callout) with best practices and a sample annotated chart script.
- Include annotated-chart example scripts showing how to place annotations, adjust visuals, and maintain clarity.
When to use it
Use the skill whenever you need to convert data into slide-ready visuals with narrative value, particularly for presentations aimed at influencing decisions (internal dashboards, executive briefings, investor decks).
What it can touch
The skill declares tools: "claude-code". It includes code blocks and Python snippets for generating charts and annotated visuals, as well as markdown templates for slide content.
Caveats
The content sticks to descriptive guidance, chart recommendations, and example code; it does not guarantee specific outcomes from visuals, and it avoids evaluative judgments beyond the stated guidance and best practices in the text.
# Data Storytelling for Slides Turn raw data into compelling visual narratives that audiences understand, remember, and act on. ## Core Principles ### 1. Data Without Story Is Just Numbers A chart without context is meaningless. Every data point should serve a narrative: "Revenue grew 40% because we launched the mobile app" is a story. "Revenue was $1.2M" is just a number. ### 2. Show the Relationship, Not Just the Data The best data slides reveal relationships: cause and effect, before and after, actual vs. target, you vs. competition. Comparison is the foundation of insight. ### 3. Maximize the Data-Ink Ratio Remove anything in a chart that isn't data or essential context. Gridlines, borders, 3D effects, excessive labels — eliminate noise. Every pixel should either inform or guide. ### 4. The Audience Should Arrive at the Insight in 5 Seconds If a viewer can't understand the key takeaway within 5 seconds, the visualization has failed. Pre-attentive attributes (color, size, position) should make the insight immediately obvious. --- ## Data Storytelling Maturity Model | Level | Chart Selection | Narrative Integration | Visual Design | Audience Impact | |-------|----------------|---
- Core Principles
- 1. Data Without Story Is Just Numbers
- 2. Show the Relationship, Not Just the Data
- 3. Maximize the Data-Ink Ratio
- 4. The Audience Should Arrive at the Insight in 5 Seconds
- Data Storytelling Maturity Model
- Chart Selection Guide
- Choosing the Right Chart Type
- Chart Decision Flowchart
- When NOT to Use Pie Charts
- Actionable Guidance
- Telling a Story with Data
- The Data-Narrative Template
- Data-Ink Ratio
What does the data-storytelling skill do?
Data Storytelling for Slides: Charts, graphs, data visualization, and narrative techniques for presenting data in presentations
How do I install it?
Run `npx skills add cosmicstack-labs/mercury-agent-skills --skill data-storytelling --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From cosmicstack-labs/mercury-agent-skills, a repository with 364 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.