Research & analysis
12,633 rows from 2,315 repositories
Going and finding out, then saying what it means.
What Research & analysis skills agree on
1,724 skill files read, by 1,213 of the 2,113 authors on this shelf whose files we hold, 2026-09-06
The middle one of the 215 measured here is ~2.0k 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
- Cite sources for every important claim47 of 1213 in 38 filesalso worded as Cite each claim's source; cite sources for every claim
- Separate facts from inferences and recommendations21 of 1213 in 12 filesalso worded as Separate verified facts from inference; Separate confirmed facts from inference
- Write findings to a markdown file19 of 1213also worded as write findings into a markdown file; Write findings to a single Markdown file
- Label every insight with a confidence level18 of 1213 in 8 filesalso worded as label every finding with confidence level; label confidence on every insight
- Read product marketing context before asking questions18 of 1213 in 8 filesalso worded as read product marketing context before starting; Read product marketing context and customer insights before starting
- Rank themes by frequency and intensity16 of 1213 in 6 filesalso worded as score themes by frequency multiplied by intensity
- Establish research mode before proceeding16 of 1213 in 6 files
- Segment survey responses by customer tier or tenure16 of 1213 in 6 filesalso worded as segment survey responses by tier or use case; Segment survey responses by customer tier or use case
- Categorize support tickets before analyzing16 of 1213 in 6 filesalso worded as categorize support tickets before analysis
- Weight research sources from the last twelve months16 of 1213 in 6 filesalso worded as weight sources from the last twelve months heavily
- Use at least five data points per segment15 of 1213 in 5 filesalso worded as use at least five data points per segment for personas
- Extract verbatim quotes for all research findings15 of 1213 in 5 filesalso worded as extract exact verbatim quotes for research
- Ask the user which deliverable format they need15 of 1213 in 5 filesalso worded as ask the user for their preferred deliverable format; ask the user for their desired deliverable format
- Ask one question at a time14 of 1213also worded as Ask questions one at a time; ask questions one at a time using AskUserQuestion
- Wait for user confirmation before proceeding14 of 1213 in 13 filesalso worded as wait for user approval before making changes; Wait for explicit human confirmation before proceeding
What they tell it not to do
- Never fabricate data24 of 1213also worded as Do not fabricate data; Do not fabricate data to fill gaps
- Do not invent details for personas20 of 1213 in 10 filesalso worded as do not invent persona details; do not invent quotes or persona details
- Do not average across segments17 of 1213 in 7 filesalso worded as Do not average across different scenarios; Do not average across different segments
- Do not name personas cutely16 of 1213 in 6 files
- Do not modify any code16 of 1213also worded as Do not modify any files; Do not modify code
- Do not treat all support tickets as equal signal15 of 1213 in 5 files
- Do not average across different churn causes15 of 1213 in 5 files
- Do not build personas from fewer than five data points15 of 1213 in 5 filesalso worded as do not create personas with fewer than five data points
- Do not skip analysis steps15 of 1213also worded as Do not skip steps; Do not skip analysis gates
- Do not skip phases14 of 1213also worded as Do not skip any analysis phases; Do not skip phases in project audits
What they expect to be installed
- git116 of 1213
- grep106 of 1213
- python389 of 1213
- WebSearch83 of 1213
- WebFetch83 of 1213
- python71 of 1213 in 67 files
- curl48 of 1213 in 44 files
- jq47 of 1213
- bash46 of 1213
- Read45 of 1213
What they ask it to produce
- Analysis report46 of 1213 in 34 filesalso worded as Performance analysis reports; analysis findings
- Markdown report40 of 1213also worded as markdown research report; Markdown file
- report.md37 of 1213also worded as research.md; findings.md
- Research report35 of 1213 in 31 filesalso worded as research findings; cited research report
- Recommendation32 of 1213 in 14 filesalso worded as Recommendations; Recommendation list
- Research synthesis report29 of 1213 in 19 filesalso worded as synthesis report; customer research synthesis report
- Research gap analysis27 of 1213 in 17 filesalso worded as Gap analysis; gap analysis report
- Executive summary26 of 1213 in 16 files
When Research & analysis authors say to reach for one
The situations these authors wrote into their own files, counted out of the same 1,213 authors, with the skills that name each one
- User requests research on a topic49 of 1213 in 47 files
- Blog discourse
- Seo geo
- Research
- Seo research
- Research collector
- Research brief
- and 6 more on this shelf
- User asks to research or compare27 of 1213 in 19 files
- Research ops
- Investigate
- and 2 more on this shelf
- Research24 of 1213
- Customer research17 of 1213 in 7 files
- Customer research
- Marketing customer research
- and 12 more on this shelf
- Voice of customer17 of 1213 in 7 files
- Customer research
- Marketing customer research
- and 12 more on this shelf
- VOC17 of 1213 in 7 files
- Customer research
- Pm feedback
- Marketing customer research
- and 12 more on this shelf
- Build personas17 of 1213 in 7 files
- Customer research
- Audience research
- Marketing customer research
- and 12 more on this shelf
- Analyze transcripts17 of 1213 in 7 files
- Customer research
- Marketing customer research
- and 12 more on this shelf
How Research & analysis skills are built
11,791 skill directories by 2,007 authors, read from their repositories’ own file trees 2026-08-05
The middle bundle among those shipping files is 3 files, 18.8 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 skill658 of 2,007 authors, 5,594 of 11,791 skills
- files ship beside it1,349 of 2,007 authors, 6,197 of 11,791 skills
- executable scripts ship inside637 of 2,007 authors, 2,659 of 11,791 skills
The folders they converge on
- references/861 of 2,007 authors, 3,249 of 11,791 skills
- scripts/541 of 2,007 authors, 2,166 of 11,791 skills
- agents/312 of 2,007 authors, 713 of 11,791 skills
- assets/227 of 2,007 authors, 664 of 11,791 skills
- examples/158 of 2,007 authors, 692 of 11,791 skills
- templates/131 of 2,007 authors, 229 of 11,791 skills
AgriciDaniel/claude-obsidian/skills/autoresearch Skill
10,848★ repoSelf-organizing AI second brain for Obsidian + Claude Code. Drop any source and Claude reads, links, and files it into one connected knowledge graph of plain Markdown you own. AI note-taking, personal knowledge management (PKM), and an open-source Notion alternative. Based on Karpathy's LLM Wiki pattern.
OpenSenseNova/SenseNova-Skills/skills/sn-da-excel-workflow Skill
4,914★ repoModular SenseNova skills for building AI-powered office assistants and productivity workflows
zebbern/claude-code-guide/skills/academic-paper-reviewer Skill
4,568★ repoClaude Code Guide - Setup, Commands, workflows, agents, skills & tips-n-tricks go from beginner to power user!
Auto Empirical Research Skills
brycewang-stanford/Auto-Empirical-Research-Skills Skill
no license3,315★🔬 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.
blazickjp/arxiv-mcp-server/io.github.blazickjp/arxiv-mcp-server MCP server
3,047★ repoSearch arXiv papers, download full text, semantic search, citation graphs, and alerts via MCP.
Weizhena/Deep-Research-skills/skills/research-codex-en/research-add-fields Skill
1,918★ repoStructured deep research skill for Claude Code/Open Code/Codex with human-in-the-loop control
a5c-ai/babysitter/.claude/skills/assimilate-popular-workflows Skill
1,665★ repoBabysitter enforces obedience on agentic workforces and enables them to manage extremely complex tasks and workflows through deterministic, hallucination-free self-orchestration
AgriciDaniel/claude-blog/brain/.raw/sources/claude-blog-skill/skills/blog-analyze Skill
1,604★ repoClaude Code blog skill suite: 30 sub-skills, 5 agents, 5-gate v1.9.0 Blog Delivery Contract, dual-optimized for Google rankings and AI citations. Active development at AI-Marketing-Hub/claude-blog (AI Marketing Hub Pro community); public releases ship here.
MCP Server for WinDbg Crash Analysis
svnscha/mcp-windbg/io.github.svnscha/mcp-windbg MCP server
1,521★ repoA Model Context Protocol server for Windows crash dump analysis using WinDbg/CDB
zanwei/design-dna Skill
1,329★Turn reference UIs (images, screenshots, URLs) into quantified Design DNA JSON—tokens, qualitative style, visual effects—then generate matching UI from your content.
ReScienceLab/opc-skills/skills/seo-geo Skill
1,282★ repoAgent Skills for Solopreneurs
CreminiAI/skillpack/skills/skillpack-creator Skill
1,197★ repoPack and deploy local AI agents for your team in minutes
AvdLee/Xcode-Build-Optimization-Agent-Skill/skills/spm-build-analysis Skill
1,195★ repoAn Agent Skill helping you to optimize Xcode incremental and clean builds by running benchmarks and optimizing build settings.
Octocode MCP - AI Context Platform
bgauryy/octocode-mcp/io.github.bgauryy/octocode-mcp MCP server
908★ repoAI code research platform. Search, analyze, and extract insights from any GitHub repository.
kostja94/marketing-skills/skills/analytics/seo/google-search-console Skill
879★ repoAgent Skills for Marketing — SEO, Social, Influencer & More. 160+ open-source skills for SEO, content, 40+ page types, paid ads, channels, and strategies. Add project context; get tailored, production-ready output. Cursor, Claude Code, OpenClaw — no lock-in.
Bhanunamikaze/Agentic-SEO-Skill Skill
831★An LLM-first SEO analysis skill for Antigravity, Codex, Claude with 16 specialized sub-skills, 10 specialist agents, and 88 optional utility scripts used as evidence collectors.
staruhub/ClaudeSkills/lab/Geek-skills-a-share-analyst Skill
672★ repo13 curated Claude Code agent skills — decks, deep research, PRDs, articles, audits. Tested like software.
Ar9av/PaperOrchestra/skills/agent-research-aggregator Skill
no license634★ repoAn automated AI research-paper writer based off Google's PaperOrchestra paper's implementation through a skills - benchmark + autoraters using any coding agent (Claude Code, Cursor, Antigravity, Cline, Aider). No API keys, no LLM SDKs.
ferdinandobons/startup-skill/startup-competitors Skill
586★ repoAI agent skills for startup validation, competitive intelligence, and planning
hanlulong/econ-writing-skill/skills/econ-write Skill
529★ repoAgent Skill that transforms AI assistants into expert economics paper writers. Synthesizes 50+ guides by Cochrane, McCloskey, Shapiro, Head, Bellemare, Goldin, Kremer. Compatible with Claude Code and OpenAI Codex.
Light0305/Light-skills/skills/light-citation Skill
521★ repoAn AI workflow skill pack for research, competitions, and innovation projects.
Optim-Agent/optim-plans/skills/analyze-and-plan Skill
494★ repoHuman-in-the-loop planning plugin for Claude and Codex: turn ideas into reviewed Markdown plans, record decisions, enforce explicit execution gates, and provide tested controller primitives for safer agent workflows.
Johell1NS/browser-search Skill
486★A skill for AI agents: search the web with SearXNG, browse with Camofox, bypass protections with CloakBrowser. Anti-hallucination by design. Self-hosted, free, unlimited.
BlockRunAI/blockrun-mcp/io.github.BlockRunAI/blockrun-mcp MCP server
426★ repoWeb search, deep research, prediction markets & crypto data for AI agents. Pay per call via x402.
ergrelet/windiff/.claude/skills/windiff-version-diff-analysis Skill
392★ repoTool that allows comparing symbol, type and syscall information of Microsoft Windows binaries across different versions of the OS, using a Web UI and/or LLMs.
tizzy916/humanities-writing-companion Skill
no license357★End-to-end humanities writing assistant — an Agent Skill (open SKILL.md format). 11 modes from Socratic research-question sharpening through AI-use disclosure. Bilingual (EN/中文), discipline-aware (literature/history/philosophy/art/religion/linguistics + cross-disciplinary). Four-layer critique, calibratable devils advocate, voice preservation.
DougTrajano/pydantic-ai-skills/examples/skills/arxiv-search Skill
357★ repoThis package implements Agent Skills (https://agentskills.io) support with progressive disclosure for Pydantic AI. Supports filesystem and programmatic skills.
K-Dense-AI/science-superpowers/skills/designing-the-analysis Skill
no license282★ repoComposable computational-science methodology skills for AI research agents — pre-registration over TDD. A science-domain reimplementation of Superpowers.
microsoft/GitHub-Copilot-for-Azure/.github/skills/analyze-skill-issues Skill
no license242★ repoGitHub Copilot for Azure
rollingSirius/equity-research-skill Skill
236★可能是最深度的 AI 投研报告 Skill：九章个股深研 + 九章财报深度分析，脚本化 DCF/EPV 与可复算估值
Designing distributed system tests
shenli/distributed-system-testing/skills/designing-distributed-system-tests Skill
226★ repoAI-agent skills for distributed-systems testing
king-of-the-grackles/reddit-research-mcp/io.github.king-of-the-grackles/reddit-research-mcp MCP server
226★ repoTurn Reddit's chaos into structured insights with full citations - MCP server for Reddit research
referodesign/refero_skill Skill
188★Research-first design skill for AI agents. 150K+ real app screens and flows via Refero MCP.
ArabelaTso/Skills-4-SE/skills/abstract-domain-explorer Skill
142★ repoA curated list of 180+ useful Claude Skills for Software Engineering and resources for customizing AI for SE workflows.
karaage0703/ai-assistant-workspace/skills/xs-arxiv Skill
136★ repoAaravKashyap12/advise-project-approach/skills/advise-project-approach Skill
134★ repoA Claude/Codex skill that makes AI agents research comparable projects, tradeoffs, costs, and failure conditions before giving build advice.
damionrashford/RivalSearchMCP/plugins/rival-search-mcp-skills/skills/rival-search-mcp Skill
119★ repoDeterministic research MCP server on FastMCP 3 — 5-engine web search, 9-platform social search, 6 academic DBs, news aggregation, entity profiles, conflict detection, document analysis. No API keys. No in-server LLM. Structured outputs for agent chaining.
BaggaT236/AI-Trading-Skills/examples/weekly-trade-strategy/skills/breadth-chart-analyst Skill
118★ repoAi agent claude trading skills for disiplined, repeatable trading workflows with a modern typescript platform
free-revalution/AIGC-Detector-Pro/.claude/skills/aigc-detector Skill
112★ repoAIGC 检测论文 AI 降率工具 - Claude Code Skill for academic paper AI detection and rewriting
infranodus/skills/infranodus-cli Skill
no license107★ repoSkills to augment LLM thinking process, integrated with InfraNodus insight generation tool
beevibe-ai/beevibe-cto/examples/claude-code-skill Skill
105★ repoArchitecture Deep Research: deep research for strategic system design decisions.
joshua-zyy/academic-paper-writer/skills/academic-citation Skill
104★ repo面向 CS / AI / ML 领域的证据驱动、分节推进的论文写作 Agent Skill。
incogbyte/android-reverse-engineering-claude-skill/plugins/android-reverse-engineering/skills/android-reverse-engineering Skill
98★ repoClaude Code skill that automates Android application reverse engineering
armelhbobdad/bmad-module-skill-forge/src/skf-analyze-source Skill
no license91★ repoA standalone BMAD module that transforms code repositories, documentation websites, and developer discourse into agentskills.io-compliant, version-pinned, provenance-backed agent skills.
foryourhealth111-pixel/research-innovation-explorer Skill
85★The automated approach leverages the cross-combination of high-quality papers from top conferences to uncover feasible research and innovation ideas. Through multi-level verification and convergence screening, it identifies research schemes that are feasible and have in-depth value.
risingwavelabs/box0 Skill
81★Open-Source Platform for Subagents and Agent Teams. Long-running, collaborative, proactive.
blacktop/ipsw-skill/ipsw Skill
75★ repoA SKILL.md for ipsw
thepushkarp/cc-gemini-plugin Skill
no license69★Gemini CLI integration for Claude Code and Codex.
mmTheBest/Research-Architect/dist/codex/skills/research-architect-audit Skill
67★ repoAn end-to-end research workflow skill for turning ideas into evidence-backed, citation-supported manuscript drafts
Agents365-ai/semanticscholar-skill/skills/semanticscholar-skill Skill
65★ repoSemantic Scholar API wrapper for agents — search 200M+ papers, traverse citations, find authors. Python, rate-limited, multi-agent.
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.
RuochenLyu/apple-health-analyst/.agents/skills/apple-health-analyst Skill
59★ repoPrivacy-first Apple Health analyzer — cross-metric insights, behavioral patterns, and health reports. Agent skill for Claude Code, Codex & Cursor.
genli-ai/market-research-skills/skills/analyst-research Skill
59★ repoTurn Claude into a disciplined research analyst — verify facts against primary sources, brief any topic, draft flagship reports. Standalone or chained; runs across LLM terminals. | 把 Claude 变成讲纪律的研究分析师：核实事实、主题简报、旗舰研报，引用一手来源、绝不造数；可单用或串联，跨 LLM 终端。
Tapetide-hq/nse-bse-indian-stock-market-data-mcp/com.tapetide/stock-research-mcp MCP server
57★ repoNSE & BSE Indian stock market data MCP — quotes, screener, financials, FII/DII flows & portfolio
Othmane-Khadri/gtm-engineer-playbook/.claude/skills/gtm-playbook/account-research-brief Skill
53★ repo10 installable Claude Code skills that turn your terminal into a GTM command center. ICP building, signal scanning, lead scoring, outreach sequences, battlecards. Built by Earleads.
ElmatadorZ/MoneyAtlas-ClaudeSkill-Agent Skill
53★Money Atlas Skill.md (Claude Skill) For AI Agent A Genesis Protocol-powered intelligence system for financial markets, macroeconomics, and geopolitics. Core Systems - Genesis Protocol (First Principle + System Thinking) - SMC Layer Engine (Quant price structure)
nik1t7n/research-craft-skill Skill
50★Agent skill for better research loops
assafkip/huntkit/skills/osint Skill
48★ repoInvestigation toolkit for Claude Code: case management, OSINT, structured analytic techniques, chain-of-custody evidence capture, and bundled threat-intel MCP servers.
WenyuChiou/research-hub/io.github.WenyuChiou/research-hub MCP server
48★ repoResearch workspace for Zotero + Obsidian + NotebookLM. Search, ingest, sync, verify briefs.
zoharbabin/web-researcher-mcp/io.github.zoharbabin/web-researcher-mcp MCP server
48★ repoYour AI research assistant that cites real sources and stays honest