This is a structured workflow for handling CSV, Excel, PDF, and image data without blowing your token budget. It enforces a safety-first approach: always probe with metadata and small samples before processing, run quality checks for missing values and duplicates, then execute Python analysis with Pandas and Matplotlib. The source includes detailed checklists for data integrity (completeness, uniqueness, consistency) and generates reports in markdown, PDF, or DOCX. It's opinionated about avoiding full data dumps and requires verification before delivery. Use this when you need repeatable exploratory analysis with guardrails, especially if you're dealing with multiple files that need cross-referencing or messy real-world data that needs cleaning first.

skills.sh installs
231
last updated
Jun 3, 2026

Install

$ npx -y skills add excelsioryy/data-analysis-skill --skill data-analysis --agent claude-code

Installs into .claude/skills of the current project.

Files

数据分析技能

支持的输入数据类型

  1. 表格形式的数据:csv、xlsx
  2. 文本中包含的数据:docx、pdf、md
  3. 图片中包含的数据:png、jpg

步骤一:数据读入与探查

为了防止 Token 溢出和执行环境卡死,智能体 严禁 直接打印全量数据或将其加载到 Prompt 上下文中,只允许将必要的数据概要加载进上下文。

1. 数据读入与保存

对不同类型的数据应使用不同的读取方法:

  • csv: 直接使用pandas库读取
  • xlsx: 使用pandas库或 \xlsx 工具读取
  • docx: 使用 \docx 工具读取,并将其中待分析的目标数据提取出来保存为csv文件
  • pdf: 使用 \pdf 工具读取,并将其中待分析的目标数据提取出来保存为csv文件
  • md: 直接读取,并将其中待分析的目标数据提取出来保存为csv文件
  • png、jpg: 使用视觉工具读取,并将其中待分析的目标数据提取出来保存为csv文件

2. 数据探查

在编写正式分析代码前,必须先执行一段轻量级的探查代码。将代码保存至工作空间。

原则:

只看样貌,不看全貌: 仅获取元数据和少量样本。

元数据优先: 必须获取列名(Columns)和数据类型(Dtypes)。

样本限制: 严格限制读取行数(推荐n=5)。

探查全部文件: 若有多份待分析的文件,需要对所有文件逐一探查,并检查文件间可能存在的关联,将具体关联方式保存为csv文档或md文件,供后续分析进行参考。

标准探查代码模式 (Python示例):


import pandas as pd

# 设定文件路径
file_path = 'target_data.csv'

Sponsors

Details

Repository
github.com/excelsioryy/data-analysis-skill
First seen
Jun 3, 2026

Recommended

1juliusbrussee avatarcavemanjuliusbrussee/cavemanUltra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels618.3Kskills.sh installs
2mattpocock avatargrill-memattpocock/skillsA relentless interview to sharpen a plan or design.1.5Mskills.sh installs
3shadcn avatarimproveshadcn/improveSurvey any codebase as a senior advisor and produce prioritized, self-contained implementation plans for OTHER models/agents to execute. Strictly read-only on s40Kskills.sh installs
4obra avatarsystematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes326.7Kskills.sh installs
5forrestchang avatarkarpathy-guidelinesforrestchang/andrej-karpathy-skillsBehavioral guidelines to reduce common LLM coding mistakes through explicit assumptions, simplicity, and verifiable success criteria.17.9Kskills.sh installs
6vercel-labs avatarfind-skillsvercel-labs/skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express int4.3Mskills.sh installs