AI Data Visualizer

核心功能：自动识别数据类型推荐最优图表组合（含折线、柱状等6类），生成带悬浮提示、明暗主题切换的响应式HTML仪表板，还提供数据统计摘要与原始数据预览。典型场景：需可视化CSV/JSON数据、制作数据分析仪表板时使用。支持文件或直接数据输入，可指定主题与标题，输出HTML文件。

openlark

热度 42.19

最近更新 2026-05-23 10:14:20

分类：数据依赖：代码执行 / 读写文件应用场景：数据处理 / 报告撰写

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add ai-data-visualizer

Metadata

name: ai-data-visualizer
description: Automatically analyze and recommend optimal chart combinations based on data characteristics, generate beautiful interactive HTML dashboards (including line charts, bar charts, scatter plots, pie charts, etc.), support dark/light theme switching, CSV and JSON input, and data statistical summaries.

AI Data Visualizer

Generate interactive HTML dashboards from CSV/JSON data with one click.

Applicable Scenarios

User provides CSV/JSON files that need visualization
User needs a data analysis dashboard
User mentions "data visualization", "generate chart", "plot", "chart"
User uploads tabular data wanting intuitive display

Quick Start

python3 "{SKILL_DIR}/scripts/generate_dashboard.py" data.csv -o dashboard.html
python3 "{SKILL_DIR}/scripts/generate_dashboard.py" data.json --json -o dashboard.html --theme dark

Features

Smart Chart Recommendation: Automatically detects column types (numeric/time/categorical/text) and recommends optimal chart combinations
6 Chart Types: Line charts (trends), bar charts (comparison), scatter plots (correlation), pie/donut charts (proportions)
Interactive HTML: Chart.js rendering, hover tooltips, dark/light theme toggle, responsive layout
Statistical Summary: Automatically calculates mean, median, min/max values
Data Table: Embedded raw data preview (limit 500 rows)

Supported Chart Selections

The script automatically selects chart strategies based on data column types. See {SKILL_DIR}/references/chart-selection.md.

CLI Options

Option	Description	Default
`input`	CSV or JSON file path	(required)
`-o, --output`	Output HTML path	`dashboard.html`
`--json`	Input is JSON format	auto-detect
`--stdin`	Read CSV from stdin	-
`--theme`	`light` or `dark`	`light`
`--title`	Dashboard title	`数据可视化仪表板`

Workflow

Receive user data — File path (CSV/JSON) or direct data content
Preprocess data — Clean and transform if necessary
Generate dashboard — Run script, output HTML
Inform user — Describe generated file and chart summary

If the user provides data content directly (rather than a file), write it to a temporary CSV file first, then call the script.

Dependencies

Python 3.7+
Chart.js 4.x (CDN loaded in HTML output)
No pip dependencies (pure standard library)

Example

Input sales.csv:

Month,Product,Revenue,Units
2024-01,A,15000,120
2024-02,A,18000,145
2024-01,B,12000,95
2024-02,B,14000,110

Output: Automatically generates a dashboard with trend line charts, category comparison bar charts, and revenue proportion pie charts.