How to use the Data Visualization Guide agent skill

What is an agent skill and what does this one do?

An agent skill is a folder containing a SKILL.md file. The file gives an AI agent specific instructions it can load when a task matches the skill's description. This particular skill is written as guidance only: it does not include scripts or tools of its own.

When the skill is active, the agent applies a checklist of visualization best practices. It considers which chart type fits your data (line for trends over time, bar for comparing categories, scatter for relationships, pie for parts of a whole with no more than 5-6 segments, heatmap for patterns in large datasets, box plot for distributions and outliers). It also applies design rules such as starting bar chart y-axes at zero unless there is a good reason not to, including units, adding legends when there are multiple series and limiting colors to 5-7.

• Chooses a chart type based on the data's characteristics and audience
• Labels titles, axes and units and adds legends for multiple series
• Uses sequential, diverging, categorical or highlight colors on purpose
• Suggests annotations, context and data sources to support the story

How do you install the data-visualization skill?

Download the skill's ZIP file from this page. Then choose one of the two installation routes below.

For Claude Code, unzip the file into your personal skills folder ~/.claude/skills/ to make it available across your projects, or into a project's .claude/skills/ folder to limit it to that project.

In the Claude apps, open Settings, go to Capabilities, and upload the skill under Skills.

• Personal install: unzip into ~/.claude/skills/
• Project install: unzip into .claude/skills/ inside your project
• Claude apps: Settings → Capabilities → Skills, then upload the ZIP

How do you trigger the skill with a natural request?

You do not need a special command. The skill is meant to activate when you ask the agent to create a chart, graph or visualization. Describe the goal and, if you have it, the data. Mentioning the audience or use case helps the agent pick a suitable style.

The skill's own examples show the kind of request that fits: a dashboard, an exploratory look at patterns, or publication-ready report charts.

• "Create an executive dashboard for sales metrics."
• "Visualize patterns in customer behavior data."
• "Generate publication-ready charts for our annual report."
• "Which chart type should I use to show monthly revenue by region?"

What does the skill need and what are its safety notes?

The skill lists no requirements. It does not run scripts and does not use the network, so it only changes the advice the agent gives you. Because it is guidance rather than code, nothing in it executes on your machine.

The skill does not give the agent access to your data. If you want charts built from real files, you need to supply the data and a tool that can run the code yourself.

• Requires: nothing listed
• Runs scripts: no
• Uses network: no
• Files included: SKILL.md only

What tips get better charts from the agent?

Give the agent the question you want answered before the chart type. The skill starts by determining the message to communicate, so a clear question produces a more focused chart.

Ask for the tools you use. The skill mentions Python libraries such as matplotlib, seaborn, plotly and altair, R's ggplot2, JavaScript options such as D3.js, Chart.js and Highcharts, and BI tools such as Tableau, Power BI and Looker. Naming one of these lets the agent match its advice to your environment.

Check the result yourself. The skill recommends verifying accuracy, but it contains no method for doing so, so confirm figures against your source before you publish.

• State the insight you want the chart to show
• Name your tool or library
• Review numbers against the original data before sharing