How do I install the Data Viz skill?

Run npx skillmds@latest add fritzprix/data-viz in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Data Viz safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Data Viz?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Data Viz free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Data Viz?

fritzprix (@fritzprix) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Data Viz by fritzprix — Agent Skill

Name: Data Viz
Author: fritzprix

Data Viz

End-user data visualization — turn CSV, Excel, or JSON into standalone HTML charts and small dashboards. Use when the user wants charts, trends, or visual reports from tabular data ("chart this CSV", "show monthly sales trend", "make a dashboard from this spreadsheet"). Not for document editing (docx/pptx) or raw format conversion (to-md). Triggers on: "차트 만들어줘", "데이터 시각화", "plot this data", "dashboard from CSV".

fritzprix f3f08fa 6 files · 17.0 KB Updated Sep 17, 2026

File contents

Data Visualization

Turn the user's tabular data into interactive HTML charts they can open in a browser.

End-user analysis and reporting skill — not LibrAgent internal analytics.

Not This Skill

Skill	Use for
to-md	Extract text from PDF/XLSX without charting
docx / pptx	Author formatted documents
deep-research	Narrative research reports
workspace-indexer	Batch markdown index for a repo

Path conventions

Replace <skill-base-dir> with this skill's absolute Base Directory.

Prerequisites

CSV and JSON work with the Python standard library.

For Excel (.xlsx):

pip install pandas openpyxl

Charts use Chart.js loaded from CDN in generated HTML (network needed when opening the file in a browser).

Workflow

1. Locate data

Use a workspace path the user provides or attachments they uploaded.

Supported inputs: .csv, .json (array of objects), .xlsx / .xls (with pandas).

2. Inspect columns

python "<skill-base-dir>/scripts/visualize.py" --action inspect --input path/to/data.csv

Review numeric_columns vs categorical_columns before charting.

3. Choose rendering mode

Mode	When	Command
`auto_chart`	User did not specify chart type	`--action auto_chart --input ... --output report.html`
`chart`	Explicit axes and type	`--action chart --chart-type line --x month --y revenue --input ...`
`dashboard`	Multiple metrics	`--action dashboard --input ... --spec dashboard.json --output dashboard.html`

Chart types: bar, line, pie, scatter.

Examples: request-patterns.md. Dashboard spec: dashboard-spec.md.

4. Present results

Tell the user:

output HTML absolute path
chart type and columns used
how to open (browser, or browser builtin navigateToUrl with file:// when allowed)

Format tips: output-format.md.

Guidelines

Inspect first on unknown files — avoid wrong axis guesses
auto_chart is a default, not magic — override with explicit chart when the user names columns
Keep data local — do not upload user data to external chart services
Large files — sample or aggregate in workspace before charting (>10k rows)
Locale — titles/labels may match user language; column names stay as in the file

Script actions

Action	Purpose
`inspect`	Schema summary
`auto_chart`	Infer bar/line/pie/scatter
`chart`	Explicit chart
`dashboard`	Multi-chart HTML

References

request-patterns.md
dashboard-spec.md
output-format.md
error-handling.md

---name: data-vizdescription: End-user data visualization — turn CSV, Excel, or JSON into standalone HTML charts and small dashboards. Use when the user wants charts, trends, or visual reports from tabular data ("chart this CSV", "show monthly sales trend", "make a dashboard from this spreadsheet"). Not for document editing (docx/pptx) or raw format conversion (to-md). Triggers on: "차트 만들어줘", "데이터 시각화", "plot this data", "dashboard from CSV".---# Data VisualizationTurn **the user's tabular data** into interactive HTML charts they can open in a browser.End-user **analysis and reporting** skill — not LibrAgent internal analytics.## Not This Skill| Skill | Use for || --- | --- || **to-md** | Extract text from PDF/XLSX without charting || **docx** / **pptx** | Author formatted documents || **deep-research** | Narrative research reports || **workspace-indexer** | Batch markdown index for a repo |## Path conventionsReplace `<skill-base-dir>` with this skill's absolute Base Directory.## PrerequisitesCSV and JSON work with the Python standard library.For Excel (`.xlsx`):```bashpip install pandas openpyxl```Charts use **Chart.js** loaded from CDN in generated HTML (network needed when opening the file in a browser).## Workflow### 1. Locate dataUse a workspace path the user provides or attachments they uploaded.Supported inputs: `.csv`, `.json` (array of objects), `.xlsx` / `.xls` (with pandas).### 2. Inspect columns```bashpython "<skill-base-dir>/scripts/visualize.py" --action inspect --input path/to/data.csv```Review `numeric_columns` vs `categorical_columns` before charting.### 3. Choose rendering mode| Mode | When | Command || --- | --- | --- || `auto_chart` | User did not specify chart type | `--action auto_chart --input ... --output report.html` || `chart` | Explicit axes and type | `--action chart --chart-type line --x month --y revenue --input ...` || `dashboard` | Multiple metrics | `--action dashboard --input ... --spec dashboard.json --output dashboard.html` |Chart types: `bar`, `line`, `pie`, `scatter`.Examples: [request-patterns.md](references/request-patterns.md). Dashboard spec: [dashboard-spec.md](references/dashboard-spec.md).### 4. Present resultsTell the user:- output HTML **absolute path**- chart type and columns used- how to open (browser, or `browser` builtin `navigateToUrl` with `file://` when allowed)Format tips: [output-format.md](references/output-format.md).## Guidelines- **Inspect first** on unknown files — avoid wrong axis guesses- **auto_chart** is a default, not magic — override with explicit `chart` when the user names columns- **Keep data local** — do not upload user data to external chart services- **Large files** — sample or aggregate in workspace before charting (>10k rows)- **Locale** — titles/labels may match user language; column names stay as in the file## Script actions| Action | Purpose || --- | --- || `inspect` | Schema summary || `auto_chart` | Infer bar/line/pie/scatter || `chart` | Explicit chart || `dashboard` | Multi-chart HTML |## References- [request-patterns.md](references/request-patterns.md)- [dashboard-spec.md](references/dashboard-spec.md)- [output-format.md](references/output-format.md)- [error-handling.md](references/error-handling.md)

Frequently asked questions