安全扫描（ClawHub）

• VirusTotal ：良性
• OpenClaw ：良性

OpenClaw 评估

技能的代码、运行时指令和所需的环境变量与基于云的图表/数据分析服务一致：它将用户文件或JSON发送到chartgen.ai进行分析/可视化并返回结果。

安装（复制给龙虾 AI）

将下方整段复制到龙虾中文库对话中，由龙虾按 SKILL.md 完成安装。

请把本段交给龙虾中文库（龙虾 AI）执行：为本机安装 OpenClaw 技能「分析数据」。简介：Chartgen技能提供三个核心功能：数据分析、数据解释和数据可视化。* *用例* * ： (1)数据分析-统计，文件...。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装：https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/chartgen-ai/analysis-data/SKILL.md
（来源：yingzhi8.cn 技能库）

SKILL.md

打开原始 SKILL.md（GitHub raw）

---
name: chartgen
description: |
  Chartgen skill providing three core functions: data analysis, data interpretation, and data visualization.
  
  **Use Cases**:
  (1) Data Analysis - Statistics, filtering, aggregation, calculation (e.g., "Calculate total sales", "Filter data greater than 100")
  (2) Data Interpretation - Trend analysis, pattern discovery, report generation (e.g., "Analyze sales trends", "Interpret data changes")
  (3) Data Visualization - Chart generation, data display (e.g., "Draw a bar chart", "Generate a pie chart")
  
  **Trigger Keywords**: analyze data, statistics, calculate, interpret trends, generate chart, visualize, plot
  
  **Prerequisites**: Set environment variable CHARTGEN_API_KEY (obtain from chartgen.ai)
metadata:
  openclaw:
    requires:
      env:
        - CHARTGEN_API_KEY
---

# ChartGen AI Chart Generator

AI-powered chart generator that creates professional visualizations through natural language. Built on ChartGen AI engine.

## Overview

Transform your data into stunning, interactive charts with simple natural language commands. No coding required - just describe the chart you want, and ChartGen AI generates it instantly.

This skill supports Text-to-Chart, Text-to-SQL, and Text-to-Data analysis. Simply provide Excel/CSV files or JSON data, describe your visualization needs in plain language, and get interactive ECharts visualizations, structured analysis reports, and AI-driven insights.

Powered by ChartGen AI engine, supporting multiple chart types including bar, line, pie, scatter, area, and more. Optimized for business analytics and data storytelling.

**API Service**: This skill uses the ChartGen API service hosted at [chartgen.ai](https://chartgen.ai). All data is sent to `https://chartgen.ai/api/platform_api/` for processing.

---

## Quick Start

### 1. Apply for an API Key

You can easily create and manage your API Key at [chartgen.ai](https://chartgen.ai). To begin with, you need to register for an account.

**Steps:**
1. Visit [chartgen.ai](https://chartgen.ai) and sign up for an account
2. Access the API management dashboard
3. Create a new API and set the credit consumption limit
4. Copy the API Key for use

### 2. Configure Environment Variable

```bash
export CHARTGEN_API_KEY="your-api-key-here"
```

### 3. Run Scripts

```bash
# Generate Chart (Text-to-Chart)
python scripts/data_visualization.py --query "Draw a bar chart of sales by region" --file sales.xlsx

# Data Analysis
python scripts/data_analysis.py --query "Calculate total sales by region" --file sales.xlsx

# Data Interpretation
python scripts/data_interpretation.py --query "Analyze sales trends" --file sales.xlsx
```

---

## Credit Rules

- Calling a single tool consumes 20 credits
- You get 200 free credits per month for free accounts
- When credits run out, you can purchase more or upgrade your account on the [chartgen.ai Billing page](https://chartgen.ai/billing)

---

## Scripts Reference

| Script | Function | Use Case |
|--------|----------|----------|
| `data_visualization.py` | Chart Generation | Text-to-Chart, create bar/line/pie/scatter charts |
| `data_analysis.py` | Data Analysis | Statistics, filtering, aggregation, calculation |
| `data_interpretation.py` | Insight Generation | Trend analysis, pattern discovery, report generation |

---

## Parameters

### Common Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--query` | Yes | Natural language query statement |
| `--file` | No | Local file path (.xlsx/.xls/.csv), mutually exclusive with --json |
| `--json` | No | JSON data (string or file path), mutually exclusive with --file |

### Visualization Specific Parameters

| Parameter | Description |
|-----------|-------------|
| `--output, -o` | Output HTML file path (defaults to /tmp/openclaw/charts/) |

---

## Data Format

### File Format

Supports `.xlsx`, `.xls`, `.csv` Excel and CSV files.

Note: Only one of --file or --json is needed. If both are provided, --file takes precedence. File types support both row-metric-column data files and column-metric-row data files.

### JSON Format

JSON data should be an array format, where each element is a row of data:

```json
[
  {"name": "Product A", "sales": 1000, "region": "East"},
  {"name": "Product B", "sales": 1500, "region": "North"},
  {"name": "Product C", "sales": 800, "region": "South"}
]
```

Or pass via file:

```bash
python scripts/data_visualization.py --query "Draw a chart" --json data.json
```

---

## Usage Examples

### Chart Generation (Text-to-Chart)

```bash
# Bar chart
python scripts/data_visualization.py --query "Draw a bar chart of sales by product" --file sales.xlsx

# Line chart
python scripts/data_visualization.py --query "Draw a line chart of sales trends" --file trends.xlsx

# Pie chart
python scripts/data_visualization.py --query "Draw a pie chart of sales by region" --file sales.xlsx

# Scatter plot
python scripts/data_visualization.py --query "Draw a scatter plot of price vs quantity" --file data.xlsx

# Save to specific path
python scripts/data_visualization.py --query "Draw a bar chart" --file data.xlsx -o /path/to/chart.html
```

### Data Analysis

```bash
# Statistical calculation
python scripts/data_analysis.py --query "Calculate total and average sales by region" --file sales.xlsx

# Data filtering
python scripts/data_analysis.py --query "Filter products with sales greater than 1000" --file sales.xlsx

# Sorting
python scripts/data_analysis.py --query "Sort by sales in descending order" --file sales.xlsx
```

### Insight Generation

```bash
# Trend analysis
python scripts/data_interpretation.py --query "Analyze monthly sales trends" --file monthly_sales.xlsx

# Anomaly detection
python scripts/data_interpretation.py --query "Find and explain anomalies in the data" --file data.xlsx

# Comprehensive interpretation
python scripts/data_interpretation.py --query "Provide a comprehensive analysis with key insights" --file report.xlsx
```

---

## Supported Chart Types

- Bar Chart / Stacked Bar Chart
- Line Chart / Area Chart
- Pie Chart / Donut Chart
- Scatter Plot
- And more...

---

## Output Description

### Chart Generation

1. **Console Output**: ECharts configuration JSON
2. **HTML File**: Interactive chart that can be opened in any browser

### Data Analysis & Insight Generation

Returns Markdown format text results, including analysis conclusions, data tables, and insights.

---

## Error Handling

Common errors and solutions:

| Error Message | Cause | Solution |
|---------------|-------|----------|
| `CHARTGEN_API_KEY not set` | Environment variable not set | `export CHARTGEN_API_KEY="your-key"` |
| `API request timeout` | Request timeout | Check network connection and retry |
| `File not found` | File does not exist | Check if file path is correct |
| `credits are insufficient` | Insufficient credits | Recharge or contact administrator |

---

## Technical Details

- **API Base URL**: `https://chartgen.ai/api/platform_api/`
- **Authentication**: Header `Authorization: <api-key>`
- **Request Format**: JSON
- **Timeout**: 60 seconds
- **Required Environment Variable**: `CHARTGEN_API_KEY`

See `scripts/chartgen_api.py` for implementation details.

---

## Privacy Notice

**Data sent to remote API**: This skill reads your provided data files (CSV/XLSX/JSON), base64-encodes them, and sends them to the ChartGen API at `https://chartgen.ai/api/platform_api/` for analysis and chart generation. Your data will leave your machine.

**Recommendations**:
- Do not upload sensitive or regulated data
- Use a dedicated API key with limited scope/credits
- Review the privacy practices at [chartgen.ai](https://chartgen.ai) before use