How to use the Data Visualization Report agent skill

What the data-report skill makes the agent do

An agent skill is a folder of instructions that an agent reads when a task matches it. This one tells the agent how to turn tabular data into a visual report page. The page is a single HTML file you can open in a browser or share.

The skill's template includes a header with a report title, a date range and a note on where the data came from. It adds a grid of three to five key metric cards, each showing a value, a year-over-year change and a small trend line. It then builds at least two charts with Chart.js or ECharts, a formatted data table with zebra striping, hover highlighting and a sticky header, three to five short insights that start with an emoji, and a collapsed methodology section.

• Parses your actual data rather than inventing numbers, according to the skill's instructions.
• Gives every chart container a fixed height so charts do not grow endlessly in the browser.
• Uses a restrained colour scheme: one main colour plus neutral greys for the chart palette.

Install the data-report skill

You can install the skill in one of three ways. Download the ZIP file from the skill's page on this site, then choose the option that fits how you use your agent.

For a personal setup, unzip the download into your skills folder at ~/.claude/skills/. For a single project, unzip it into the .claude/skills/ folder inside that project. Either way, the data-report folder should contain SKILL.md.

If you use the Claude apps, you can instead upload the ZIP under Settings → Capabilities → Skills.

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

Trigger the skill with a natural request

You do not need a special command. Attach or point the agent to your data file and describe the report you want in ordinary language. The skill is designed for data files in CSV, Excel or JSON format, and the agent must be able to read the file you provide.

The skill's description is about turning CSV, Excel or JSON data into a visual report page, so requests that name the data and the outcome work best. Mention the time period, the audience or the metrics if they matter to you.

• Turn this monthly export into a visual report page with charts and a data table.
• Make a weekly summary of this spreadsheet as a shareable HTML page.
• Build a styled report from this JSON file with key metrics and two charts.
• Use the data-report skill on this CSV and include a methodology note.

What the sample report shows

The skill ships with an example data file and a matching example report. The sample is nine months of user growth data with columns for month, active users, new signups, paying users and churn rate. It is rendered as a KPI card grid, Chart.js charts and a table.

Use the example to see the layout before you run the skill on your own data. The example's insights are generated text drawn from that dataset, so check them against your own numbers before relying on them.

• Example files: example.html (the rendered report), example.md (the data and sample write-up) and SKILL.md (the instructions).
• The example format is CSV.

Requirements and safety notes

The skill needs an agent that can read the CSV, Excel or JSON file you give it. The rendered page loads Chart.js or ECharts from the jsDelivr CDN, so the page needs internet access when you open it. The skill does not run scripts, but it does use the network.

Be careful with sensitive or confidential financial data. The agent handles your data, and the page pulls chart libraries from a public CDN. Review what leaves your machine before you send a file to an agent or an online service.

Do not use the output for decisions that need verified statistics. The insights are generated text, and the skill asks only for a methodology note. It does not validate the figures.

• Requires: an agent that can read CSV, Excel or JSON; internet access in the rendered page for jsDelivr CDN.
• Runs scripts: no.
• Uses network: yes.
• Best for weekly or monthly summaries, shareable pages, and a starting layout you can refine.
• Not for confidential data sent to an online service or for decisions that need verified statistics.

Tips for better reports

Give the agent a clean export with clear column names, such as month, active_users or churn_rate. Consistent date formats make the time range and trend lines easier to produce.

Check the chart and table against your source before sharing. Then edit the generated HTML to adjust wording, colours or layout. The skill is a starting layout, so refinement is expected.

• Use one table per file with a header row.
• Ask for a methodology note to be kept visible to readers.
• Keep chart containers at fixed heights if you edit the HTML, as the skill's template does.