data-analysis

Automate end-to-end data analysis workflows for tables, documents, and images.

1|2|Updated Feb 28, 2026

Search Tags:#multimodal #reporting #visualization #pandas #data-analysis #eda

Ready to run

npx skills add https://github.com/zhangdszq/vk-skills --skill data-analysis-zhangdszq

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: data-analysis
Source: https://github.com/zhangdszq/vk-skills/tree/main/data-analysis
Command: npx skills add https://github.com/zhangdszq/vk-skills --skill data-analysis-zhangdszq

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

解决跨来源数据的探索性分析与洞察产出难题——从表格、文档和图像中提取有用信息、进行质量检查、执行分析并生成可分享的报告。

Core Features & Use Cases

安全探查与样本化读取，避免全量加载以保护敏感数据。
多源数据处理：CSV/Excel、PDF/DOCX/MD 和图片中的数据提取、清洗、聚合与统计。
自动化分析与可视化：生成描述性统计、图表与洞察报告，支持重复执行与版本化。
Use Case: 研发日志的质量检查、销售数据的多模态分析，以及运营报告的自动化输出。

Quick Start

提供一个样本数据集，运行分析流程以获得初步洞察和可下载的报告。

Frequently Asked Questions about data-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I automate exploratory data analysis and visualization for CSV and Excel files?▼

Automating exploratory data analysis for CSV and Excel files is done by extracting, cleaning, and analyzing data with Pandas, then generating descriptive statistics and visualizations using Matplotlib and Seaborn. It supports sample-based probing to avoid full data loads and produces reproducible insight reports.

Can I extract and analyze data from PDF documents and images?▼

You can extract and analyze data from PDF documents and images through multi-modal data processing. The workflow applies data extraction and cleaning techniques across PDF, DOCX, MD, and image sources, then aggregates the results for statistical analysis and visualization.

What is the best way to generate reproducible data insights reports from multiple sources?▼

Generating reproducible data insights reports from multiple sources involves running an end-to-end workflow that performs safety checks, validates data quality, and applies versioned reporting steps. It uses Pandas for aggregation and Matplotlib or Seaborn for visualizations to ensure consistent outputs.

Does this data analysis workflow require loading my entire dataset at once?▼

This data analysis workflow does not require loading your entire dataset at once. It implements safety checks and sample-based probing to protect sensitive data, preventing full-volume loading while still executing exploratory data analysis and generating reliable insights.

How do I check data quality before running a full analysis on my tables?▼

Checking data quality before running a full analysis is handled through automated safety checks and sample-based probing. The workflow validates your tables, cleans extracted information, and ensures data integrity before proceeding to statistical aggregation and visualization.