Data storytelling
Portable agent workflow framework for Codex, Claude Code, and OpenCode—shared project context, focused skills, independent review, and safe updates.
npx -y skills add K95M65/AI_ONBOARD --skill data-storytellingAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 19 days oldThe repository was created 19 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Turn analysis into a clear, persuasive narrative for a decision-maker — message hierarchy, executive structure, and honest communication. Use when presenting findings, writing a brief or memo, or building a deck from data. Pairs with dataviz.
SKILL.md
1.8 KB, as published. Nobody here has run it
Data storytelling
Analysis that isn't communicated doesn't change a decision. The last mile of
competitive-intel; pairs with dataviz for the visuals.
Principles
- Lead with the answer (BLUF). Bottom line up front — a busy reader decides in the first sentence whether to keep going. Don't narrate your research journey.
- Pyramid structure. Answer → key supporting arguments → evidence. Not chronological.
- One message per exhibit. A chart's takeaway is its title (see
dataviz). - Tailor to altitude. Executives want the implication and the ask; practitioners want the method. Same facts, different emphasis.
Structure of a brief / memo
- Recommendation — the decision and your answer, in 1–3 sentences.
- Why — the 2–4 key reasons, each with its evidence.
- Risks / what would change the answer.
- The ask — the specific decision or next step you need.
Communication rules
- Say the "so what," not just the "what." Every fact earns its place by supporting the recommendation; cut the rest to an appendix.
- Calibrate confidence in words. "Evidence suggests" ≠ "confirms." Don't launder inference as fact.
- Length is not rigor. Cut ruthlessly; depth goes in the appendix.
- Visuals: a table for precise values, a chart for the pattern — never a chart to look busy.
Ethics
Represent uncertainty and source quality honestly. Never cherry-pick data to fit a predetermined conclusion — the fastest way to lose a decision-maker's trust for good.