Data analysis
1,579 rows from 297 repositories
Getting an answer out of a dataset.
What Data analysis skills agree on
219 skill files read, by 230 of the 242 authors on this shelf whose files we hold, 2026-09-06
The middle one of the 37 measured here is ~1.4k tokens long, counted with cl100k_base
Counted by distinct author, so one author publishing three of these counts once. Where a claim sits in fewer files than authors, that is said: a claim held by forty authors across three files is one file people copied, not forty people who agreed. Near-identical wordings are grouped and the other wordings are shown, so the grouping is yours to check.
What they tell the agent to do
- Propose a regression test for each fixed bug16 of 230 in 12 filesalso worded as Write a regression test for every bug fixed; Convert every production bug into a regression test
- Name tests after the bug they prevent14 of 230 in 10 files
- Test the API response shape, not the implementation14 of 230 in 10 filesalso worded as Test API response shape, not implementation; test the response shape, not the implementation
- Run the test suite before any code review14 of 230 in 10 filesalso worded as Run the test suite and build before agent review; run tests and build before any code review
- Force sandbox mode in the test setup14 of 230 in 10 filesalso worded as Force sandbox mode in test setup; Force sandbox mode for database-free API testing
- Write regression tests only for bugs already found14 of 230 in 10 filesalso worded as Write a regression test reproducing the bug; Write tests for bugs found, not working code
- Assert sandbox and production paths return the same shape14 of 230 in 10 filesalso worded as assert sandbox and production responses share the same shape; Assert sandbox and production paths return the same fields
- Clear stale state when setting an error13 of 230 in 9 filesalso worded as Clear related state when an error occurs; clear stale state when an error occurs
- Keep the whole test suite under one second10 of 230 in 6 filesalso worded as Keep tests under one second; Keep tests under one second total
- Run the build type check before code review10 of 230 in 6 filesalso worded as Run the build before any code review; Run the build before code review
- Use vectorized operations instead of row iteration9 of 230 in 6 filesalso worded as Use vectorized operations instead of loops; vectorize operations instead of using loops
- Start bar chart Y-axes at zero8 of 230 in 7 filesalso worded as Start bar chart axes at zero; Start y-axis at zero for bar charts
- Match chart type to the data relationship8 of 230 in 7 filesalso worded as Match chart type to data and message; Match each chart type to its data relationship
- Load the matching reference file for the task8 of 230 in 5 filesalso worded as Load the matching format reference file; load the matching topic reference file
- Use colorblind-safe palettes8 of 230 in 7 filesalso worded as Use colorblind-friendly palettes; Use colorblind-safe palettes with redundant encoding
What they tell it not to do
- Never skip sandbox path testing14 of 230 in 10 filesalso worded as skip sandbox path testing; Skipping sandbox path testing
- Never write integration tests when unit tests suffice14 of 230 in 10 filesalso worded as write integration tests when unit tests suffice; Writing integration tests when unit tests suffice
- Never trust AI self-review as a substitute for automated tests14 of 230 in 10 filesalso worded as trust AI self-review instead of automated tests; Do not trust AI self-review over automated tests
- Never write tests for code that has had no bugs14 of 230 in 10 filesalso worded as Never write tests for code that never had a bug; write tests for code that has never had a bug
- Never aim for coverage percentage instead of regression prevention14 of 230 in 10 filesalso worded as Aiming for coverage percentage instead of regression prevention; Never aim for coverage percentage over regression prevention
- Never rely on color alone to distinguish series9 of 230 in 8 filesalso worded as Relying on color alone; Never rely on red/green color distinction alone
- Never claim causation from correlation9 of 230also worded as Never claim causation from correlation alone; Do not treat correlation as causation
- Avoid pie charts with six or more categories8 of 230 in 7 filesalso worded as Never use pie charts with more than five slices; Avoid pie charts with more than five categories
- Do not ignore SettingWithCopyWarning7 of 230 in 5 filesalso worded as don't ignore SettingWithCopyWarning; Do not ignore convergence warnings
- Do not iterate over DataFrame rows7 of 230 in 5 filesalso worded as Don't iterate over DataFrame rows; don't iterate DataFrame rows with .iterrows()
What they expect to be installed
- pandas43 of 230 in 35 files
- matplotlib24 of 230 in 23 files
- numpy23 of 230 in 17 files
- seaborn19 of 230 in 18 files
- npm run build16 of 230 in 12 files
- Vitest14 of 230 in 10 files
- npm run test14 of 230 in 10 files
- plotly13 of 230 in 12 files
- python13 of 230 in 11 files
- uv10 of 230
What they ask it to produce
- Regression test for each fixed bug15 of 230 in 11 filesalso worded as Regression test reproducing the bug; a regression test for each fixed bug
- Interactive data visualizations7 of 230 in 3 filesalso worded as data visualization
- Comments explaining complex transformations6 of 230 in 4 files
- Code with vectorized operations and proper indexing6 of 230 in 4 filesalso worded as vectorized code with proper indexing
- Memory and performance notes for large datasets6 of 230 in 4 filesalso worded as Memory or performance considerations for large datasets
- Validation checks for dtypes, nulls, and shapes6 of 230 in 4 filesalso worded as data validation checks for dtypes, nulls, and shapes; Data validation checks for dtypes, nulls, shapes
- Excel file with zero formula errors5 of 230 in 4 filesalso worded as Excel workbook with zero formula errors
- D3 charts5 of 230 in 1 file
When Data analysis authors say to reach for one
The situations these authors wrote into their own files, counted out of the same 230 authors, with the skills that name each one
- AI agent has modified API routes or backend logic13 of 230 in 9 files
- Ai regression testing
- and 9 more on this shelf
- A bug is found and fixed12 of 230 in 8 files
- Ai regression testing
- and 9 more on this shelf
- Project has a sandbox or mock mode12 of 230 in 8 files
- Ai regression testing
- and 9 more on this shelf
- A bug-check or review command is being run11 of 230 in 7 files
- Ai regression testing
- and 1 more on this shelf
- Multiple code paths such as sandbox versus production exist9 of 230 in 5 files
- Ai regression testing
- and 7 more on this shelf
- User asks to explore, analyze, or summarize a data file9 of 230 in 8 files
- Exploratory data analysis
- Data analysis
- and 5 more on this shelf
- User requests a zoomable treemap6 of 230 in 2 files
How Data analysis skills are built
1,511 skill directories by 280 authors, read from their repositories’ own file trees 2026-08-05
The middle bundle among those shipping files is 2 files, 12.2 KB beside SKILL.md
Counted by distinct author, same as above, so one author publishing forty template copies counts once. SKILL.md itself is not counted as a file, so a single-file skill is one where that file is the whole skill.
The shape
- SKILL.md is the whole skill117 of 280 authors, 867 of 1,511 skills
- files ship beside it163 of 280 authors, 644 of 1,511 skills
- executable scripts ship inside94 of 280 authors, 295 of 1,511 skills
The folders they converge on
- references/85 of 280 authors, 252 of 1,511 skills
- scripts/67 of 280 authors, 174 of 1,511 skills
- assets/29 of 280 authors, 73 of 1,511 skills
- agents/20 of 280 authors, 33 of 1,511 skills
- examples/17 of 280 authors, 123 of 1,511 skills
- evals/8 of 280 authors, 27 of 1,511 skills
OpenSenseNova/SenseNova-Skills/skills/sn-da-excel-workflow Skill
4,914★ repoModular SenseNova skills for building AI-powered office assistants and productivity workflows
Mckinsey style visualization skill
kgraph57/mckinsey-style-visualization-skill Skill
65★Agent Skill that turns messy notes into rendered strategy-consulting visuals, with SVG examples and validation.
oogunbiyi21/stats-compass-mcp/io.github.oogunbiyi21/stats-compass MCP server
17★ repo50+ pandas-powered tools for data loading, cleaning, visualization, and ML workflows
alphaparkinc/genpark-growth-data-analyst Skill
no license9★GenPark skill: genpark-growth-data-analyst based on GrowthGPT research
ChengJiale150/jupyter-mcp-server/io.github.ChengJiale150/jupyter-mcp-server MCP server
9★ repoA powerful MCP server for AI-driven Jupyter Notebook management and execution
droxer/Synapse/backend/agent/skills/bundled/data-analysis Skill
5★ repoAn open-source AI agent platform that does the work for you.
kgraph57/hbr-style-visualization Skill
3★Claude Code skill for HBR-style editorial charts and business visualizations.
Agents365-ai/seurat-skill Skill
3★Seurat v5 single-cell analysis skill for Claude Code, OpenClaw, and Codex
kernelshreyak/agentic-data-grokker/.codex/skills/data-analysis-workflow Skill
no license1★ repoCodex/Claude Code-native workspace for analyzing local data files and saving reproducible notes, SQL, and charts.
AmberLee2427/unity-skills/skills/unity-jupyter Skill
1★ repoA specialized Agent Skill designed for researchers and students using the Ohio State University (OSU) Arts and Sciences (ASC), Unity HPC cluster
k4thir/smart-eda Skill
1★Adaptive exploratory data analysis as a Claude Skill — interactive dashboard + detailed report from any tabular file.
genapohub/data-analyst-guide Skill
1★AI Agent Skill:数据分析师方案产出指南,自动识别5类场景,按清单产出指标体系/埋点方案/BI看板/AB实验/数据仓库等完整交付物。兼容 WorkBuddy / Codex / Trae / Cursor。
Mission control 3d visual regression
MN755/Codex-Mission_Control/.codex/plugins/mission-control/skills/mission-control-3d-visual-regression Skill
1★ repoAn Open Source Platform For Building Projects With OpenAI's Codex, Claude CLI, And Other Coding LLMs Using Multi-Agent Swarms And Customizable Workflows. Runs On Windows, MacOS, And Linux.
alexngai/skill-tree/docs/demo/skills/git-bisect-regression Skill
0★ repoVersion control and evolution for AI agent skills — versioned storage, dynamic serving/loadouts, multi-agent git sync, and cross-repo federation.
SWEStash/swe-workflow-skills/plugins/ai/skills/ai-evaluation Skill
0★ repoA comprehensive SWE workflow, encoded. Might be useful to you too.
wyc79/research-viz-skill/research-viz Skill
0★ repoAI agent skill for research visualizations
Alpha-Park/genpark-growth-data-analyst Skill
no license0★GenPark skill: genpark-growth-data-analyst based on GrowthGPT research
xuluoforcainiao/analytics-data-analysis/analytics-data-analysis Skill
no license0★ repoImplement analytics, data analysis, and visualization best practices using Python, Jupyter, and mode
statechange/monoskill/.agents/skills/data-visualization Skill
0★ repoCompile distributed agent skills into one provenance-aware router skill
yrris/pro-agent/cognition/runtime/skills/chart-visualization Skill
no license23★ repo多智能体应用与任务执行平台
Galius5136/databricks-spark-3.5-cert-prep/skills/pandas-on-spark Skill
12★ repoStudy system for the Databricks Certified Associate Developer for Apache Spark 3.5 exam. 5 interconnected Claude Code skills covering all 7 exam sections, with sources linked to Apache Spark 3.5 docs and Damji's Learning Spark 2nd Edition.
Marazii/research-co-pilot/skills/data-analysis Skill
9★ repoComprehensive research co-pilot for Claude Code and claude.ai — 14 skills covering literature review, methodology, ethics review, data analysis, qualitative coding, brainstorming, manuscript drafting, replication design, grant writing, talk building, citations, survey design, and peer review. A peer collaborator, not a subservient assistant.
pranav8494/team-of-agents/skills/data-analyst Skill
7★ repoA team of agents to support SDLC of a project.
sciskillhub/sciskill/community/sciskillhub Skill
no license5★ repoagent skills for science
Data visualization and figures
jjfroehlich/agent-skills-for-academic-research/skills/data-visualization-and-figures Skill
1★ repoAgent skills for academia research: writing, dataviz, publishing, communication, career, and more.
UtkarshDubeyGIT/agent-skills/.agents/skills/data-visualization Skill
no license1★ repoRepository of curated and relevant agent skills to enhance Agentic Development
ksed8/cc-loopkit/.claude/skills/data Skill
1★ repoLoopkit for Claude Code - guardrails, hooks, and two loop modes that won't let a task "finish" until it's actually done.
Bio data visualization ggplot2 fundamentals
yigityildiz0/scientific-agent-skills/skills/common/bio-data-visualization-ggplot2-fundamentals Skill
no license1★ repoBilingual, searchable Agent Skills for reproducible scientific research across Claude Code, OpenAI Codex, and OpenCode.
SkillMedev/data-science-ml/skills/eda-playbook Skill
0★ repoRigorous skills for every stage of the ML lifecycle, from EDA to production monitoring.
WindcleaverDev/regkit/.agents/skills/linear-regression Skill
0★ repoStatistically rigorous regression skills for Claude — Python computes, the model narrates.
BenGardiner/ipynb-agent-skills/jupyter-notebook-linter Skill
0★ reposkills for your agent to work with jupyter notebooks
weakened-fixedinvestmenttrust229/pm-skills-ko/pm-data-analytics/skills/cohort-analysis Skill
0★ repoProvide structured AI-powered PM skills in Korean to improve product decisions from discovery to growth with Claude Code and Cowork integration.
Lukk17/agent-standards/.agents/skills/ai-regression-testing Skill
0★ repoOne git checkout drops a shared AI coding setup (skills, subagents, MCP servers, OpenSpec scaffolding) into any project, across Claude Code, Kilo, OpenCode, Codex, and Copilot.
palmiro72-coder/palmiros-claude/plugins/palmiros-clinical-br/skills/clinical-data-analysis Skill
0★ repoMarketplace de plugins do Claude Code — 14 skills agrupadas em 5 plugins: operações clínicas BR (endo + derm), revenue cycle US, pesquisa científica (harmonia sexagesimal + patentes + mineração), criação visual via Gemini e SecOps/infra.
inosaint/data-vizard/extensions/data-vizard/skills/data-analyst Skill
0★ repoA working repo for the NPM package 'data-vizard', a suite of skills to create data visualisations.
nexu-io/open-design/skills/d3-visualization Skill
85,608★ repo🎨 The open-source Claude Design alternative. 🖥️ Local-first desktop app. 🖼️ Your coding agent becomes the design engine: prototypes, landing pages, dashboards, slides, images & video — real files, HTML/PDF/PPTX/MP4 export. 🤖 Claude Code / Codex / Cursor / Gemini / OpenCode / Qwen & 20+ CLIs via BYOK.
ComposioHQ/awesome-claude-skills/document-skills/xlsx Skill
no license72,074★ repoA curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows
K-Dense-AI/scientific-agent-skills/skills/exploratory-data-analysis Skill
33,548★ repoTurn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 170,000+ scientists worldwide. 154 ready-to-use skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.
K-Dense-AI/scientific-agent-skills/skills/scientific-visualization Skill
33,548★ repoTurn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 170,000+ scientists worldwide. 154 ready-to-use skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.
K-Dense-AI/scientific-agent-skills/skills/seaborn Skill
33,548★ repoTurn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 170,000+ scientists worldwide. 154 ready-to-use skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.
K-Dense-AI/scientific-agent-skills/skills/statistical-analysis Skill
33,548★ repoTurn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 170,000+ scientists worldwide. 154 ready-to-use skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.
K-Dense-AI/scientific-agent-skills/skills/statistical-power Skill
33,548★ repoTurn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 170,000+ scientists worldwide. 154 ready-to-use skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.
phuryn/pm-skills/pm-data-analytics/skills/cohort-analysis Skill
25,257★ repoPM Skills Marketplace: 100+ agentic skills, commands, and plugins — from discovery to strategy, execution, launch, and growth.
alirezarezvani/claude-skills/engineering-team/skills/senior-data-scientist Skill
24,478★ repo345 Claude Code skills & agent skills & plugins (30+ Agents, 70+ custom commands, 330+ skills, customizable references, scripts)for Claude Code, Codex, Gemini CLI, Cursor, and 8 more coding agents — engineering, marketing, product, compliance, C-level advisory, research, business operations, commercial & finance, and your daily productivity skills.
Jeffallan/claude-skills/skills/pandas-pro Skill
10,966★ repo66 Specialized Skills for Full-Stack Developers. Transform Claude Code into your expert pair programmer.
ThinkInAIXYZ/deepchat/resources/skills/xlsx Skill
6,220★ repo🐬DeepChat - A smart assistant that connects powerful AI to your personal world
OpenSenseNova/SenseNova-Skills/skills/sn-da-excel-workflow/capability/excel-data-visualization/bar-chart-visualization Skill
4,914★ repoModular SenseNova skills for building AI-powered office assistants and productivity workflows
OpenSenseNova/SenseNova-Skills/skills/sn-da-excel-workflow/capability/excel-data-visualization/line-chart-visualization Skill
4,914★ repoModular SenseNova skills for building AI-powered office assistants and productivity workflows
OpenSenseNova/SenseNova-Skills/skills/sn-da-excel-workflow/capability/excel-data-visualization/pie-chart-visualization Skill
4,914★ repoModular SenseNova skills for building AI-powered office assistants and productivity workflows
OpenSenseNova/SenseNova-Skills/skills/sn-da-excel-workflow/capability/excel-data-visualization/stacked-chart-visualization Skill
4,914★ repoModular SenseNova skills for building AI-powered office assistants and productivity workflows
OpenSenseNova/SenseNova-Skills/skills/sn-da-excel-workflow/capability/excel-result-export/report-generation-export Skill
4,914★ repoModular SenseNova skills for building AI-powered office assistants and productivity workflows
zebbern/claude-code-guide/skills/regression-modeler Skill
4,568★ repoClaude Code Guide - Setup, Commands, workflows, agents, skills & tips-n-tricks go from beginner to power user!
brycewang-stanford/Auto-Empirical-Research-Skills/plugins/empirical-analysis-python/skills/pipeline Skill
no license3,315★ repo🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库,覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文,并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.
00.1 Full empirical analysis skill Python
brycewang-stanford/Auto-Empirical-Research-Skills/skills/00.1-Full-empirical-analysis-skill_Python Skill
no license3,315★ repo🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库,覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文,并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.
brycewang-stanford/Auto-Empirical-Research-Skills/skills/09-meleantonio-awesome-econ-ai-stuff/_skills/analysis/python-panel-data Skill
no license3,315★ repo🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库,覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文,并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.
brycewang-stanford/Auto-Empirical-Research-Skills/skills/09-meleantonio-awesome-econ-ai-stuff/_skills/analysis/stata-regression Skill
no license3,315★ repo🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库,覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文,并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.
brycewang-stanford/Auto-Empirical-Research-Skills/skills/09-meleantonio-awesome-econ-ai-stuff/_skills/communication/econ-visualization Skill
no license3,315★ repo🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库,覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文,并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.
brycewang-stanford/Auto-Empirical-Research-Skills/skills/12-pedrohcgs-claude-code-my-workflow/dot-claude/skills/data-analysis Skill
no license3,315★ repo🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库,覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文,并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.
brycewang-stanford/Auto-Empirical-Research-Skills/skills/14-luischanci-claude-code-research-starter/dot-claude/skills/data-analysis Skill
no license3,315★ repo🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库,覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文,并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.
brycewang-stanford/Auto-Empirical-Research-Skills/skills/15-Felpix-Studios-social-science-research/skills/data-analysis Skill
no license3,315★ repo🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库,覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文,并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.