data-analysis-partner

What This Skill Does

The data-analysis-partner skill for OpenClaw is a powerful, automated toolkit designed to transform raw CSV and Excel files into professional, interactive data insight reports. It bridges the gap between complex raw data and actionable business intelligence by leveraging pandas and NumPy for heavy lifting, and ECharts for high-quality visualization. When you provide a file, this skill automatically performs data cleansing, statistical profiling, and pattern recognition, culminating in a self-contained HTML document that includes executive summaries, detailed descriptive statistics, and interactive charts.

Installation

To integrate this skill, use the command: clawhub install openclaw/skills/skills/1992huanghai/data-analysis-partner. Upon the first execution, the system will attempt to provision an isolated Python environment and install necessary dependencies (pandas, openpyxl, xlrd). If your system restricts automatic installation, you can manually setup the virtual environment at ~/.openclaw/skills/data-analysis-partner/.venv and install the requirements via pip.

Use Cases

This skill is ideal for analysts, project managers, and business users who need rapid insights without writing custom code. Use it for:

• Exploratory Data Analysis (EDA) to find patterns, anomalies, or correlations in sales or usage data.
• Generating automated, shareable BI reports for stakeholders.
• Performing diagnostic analysis on customer behavior or operational log data.
• Quickly auditing dataset integrity, including identifying missing values and statistical outliers.

Example Prompts

1. "I've uploaded my Q3 customer churn data. Can you analyze this file and build a report highlighting the primary churn nodes in our user funnel?"
2. "Here is the marketing spreadsheet. I need a visual report showing the correlation between ad spend and lead conversion rates across different regions."
3. "Please analyze this sensor data file. I need to know if there are any abnormal spikes or recurring trends in the error rates over the last month."

Tips & Limitations

• Performance: For datasets larger than 100MB, please allow 1-2 minutes for processing. For datasets exceeding 50,000 rows, the skill performs statistical sampling to maintain optimal response times while noting the original record count.
• Offline Usage: The generated reports use a CDN to load ECharts by default. To enable permanent offline viewing, simply open the report once while connected to the internet to allow your browser to cache the library components.
• Privacy: Data content remains local to your environment. Only the visualization framework is fetched from external sources.